# Compare AI PC builds side by side | What Can I Run AI? > Compare local AI PC builds: price, usable VRAM, $ per GB, bandwidth, estimated tok/s, $ per tok/s, power, PSU headroom and fit warnings. Source: https://whatcanirunai.vercel.app/builds/compare/ [Builds](https://whatcanirunai.vercel.app/builds/) # Compare AI PC builds side by side Put up to three builds next to each other: total price, usable memory, $ per GB, bandwidth, estimated tokens per second, $ per tok/s, power, running cost and fit warnings, plus every model's fit. Pick recipes, or paste a share link from the [planner](https://whatcanirunai.vercel.app/planner/). Builds Build Aor paste a planner link Build Bor paste a planner link Build Cor paste a planner link WorkloadModel Quant Electricity $/kWh Copy compare link Cheapest: **C** (RTX 5090 single-GPU build (32 GB)) at $4,533\. Most memory: **A** (Dual used RTX 3090 build (48 GB) for 32B to 70B models) with 48.0 GB usable. Fastest on Qwen3-32B at Q4: **C** (RTX 5090 single-GPU build (32 GB)) (54–77 tok/s, estimate). Best $ per tok/s: **C** (RTX 5090 single-GPU build (32 GB)). ## Head to head __Build comparison, best value in each row marked Best__ | Metric | A: Dual used RTX 3090 build (48 GB) for 32B to 70B models | B: Dual RX 7900 XTX build (48 GB VRAM) | C: RTX 5090 single-GPU build (32 GB) | | ----------------------------- | --------------------------------------------------------- | -------------------------------------------------- | ------------------------------------------------------------------------------- | | Total price | $7,216 | $5,134 | $4,533 Best | | Usable memory | 48.0 GB Best | 48.0 GB Best | 32.0 GB | | $ per usable GB | $150 | $107 Best | $142 | | Memory bandwidth | 936 GB/s per card | 960 GB/s per card | 1,792 GB/s Best | | Qwen3-32B at Q4 | Fits | Fits | Fits | | Est. tok/s (Qwen3-32B) | 28–40 | 29–41 | 54–77 Best | | $ per tok/s | $211 | $147 | $69 Best | | Models that fit at Q4 | 511 Best | 511 Best | 504 +3 tight | | Biggest model that fits at Q4 | [AI21-Jamba2-Mini](https://whatcanirunai.vercel.app/models/ai21-jamba2-mini/) Best | [AI21-Jamba2-Mini](https://whatcanirunai.vercel.app/models/ai21-jamba2-mini/) Best | [IQuest-Coder-V1-40B-Loop-Instruct](https://whatcanirunai.vercel.app/models/iquest-coder-v1-40b-loop-instruct/) | | Load power | 1020 W | 1030 W | 790 W Best | | Cost per hour at $0.18/kWh | $0.18 | $0.19 | $0.14 Best | | PSU needed / headroom | 1350 W headroom 174 W | 1350 W headroom 161 W | 1050 W headroom 173 W | | Warnings | 1 | 1 | 0 Best | | GPU condition | Used | New | New | "Best" marks the leading build in each row (lowest price, power and $ figures; highest memory, bandwidth and speed). Tok/s = bandwidth ÷ weight size × an efficiency band calibrated on llama.cpp scoreboards ([how we estimate speed](https://whatcanirunai.vercel.app/method/#speed)); mixture-of-experts models show only an upper bound, so they get no $ per tok/s. Multi-GPU builds add memory, but speed is about one card's. Load power is a ceiling (GPU board power + CPU TDP + 150 W); "—" means a spec we could not source. ## Parts __Parts in each build__ | Part | A: Dual used RTX 3090 build (48 GB) for 32B to 70B models | B: Dual RX 7900 XTX build (48 GB VRAM) | C: RTX 5090 single-GPU build (32 GB) | | ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | GPU | 2× Used RTX 3090 24GB (ASUS TUF, "Like New")$4,080 · [used listing](https://www.newegg.com/asus-tuf-gaming-tuf-rtx3090-o24g-gaming-geforce-rtx-3090-24gb-graphics-card-triple-fans/p/N82E16814126454) | 2× Radeon RX 7900 XTX$1,998 · [list price (MSRP)](https://en.wikipedia.org/wiki/Radeon%5FRX%5F7000%5Fseries) | GeForce RTX 5090$1,999 · [list price (MSRP)](https://www.nvidia.com/en-us/geforce/graphics-cards/50-series/rtx-5090/) | | CPU | AMD Ryzen 9 9950X$539 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16819113841) | AMD Ryzen 9 9950X$539 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16819113841) | AMD Ryzen 7 9700X$339 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16819113843) | | Motherboard | ASUS ProArt X870E-CREATOR WIFI$512 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16813119688) | ASUS ProArt X870E-CREATOR WIFI$512 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16813119688) | ASUS TUF GAMING B650E-PLUS WIFI$150 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16813119739) | | RAM | Corsair Vengeance 64GB (2x32) DDR5-6000 CL30$1,275 · [Corsair store price](https://www.corsair.com/us/en/p/memory/cmk64gx5m2b6000c30/vengeance-64gb-2x32gb-ddr5-dram-6000mt-s-cl30-memory-kit-black-cmk64gx5m2b6000c30) | Corsair Vengeance 64GB (2x32) DDR5-6000 CL30$1,275 · [Corsair store price](https://www.corsair.com/us/en/p/memory/cmk64gx5m2b6000c30/vengeance-64gb-2x32gb-ddr5-dram-6000mt-s-cl30-memory-kit-black-cmk64gx5m2b6000c30) | Corsair Vengeance 64GB (2x32) DDR5-6000 CL30$1,275 · [Corsair store price](https://www.corsair.com/us/en/p/memory/cmk64gx5m2b6000c30/vengeance-64gb-2x32gb-ddr5-dram-6000mt-s-cl30-memory-kit-black-cmk64gx5m2b6000c30) | | Power supply | Corsair HX1500i$320 · [Corsair store price](https://www.corsair.com/us/en/p/psu/cp-9020309-na/hx1500i-fully-modular-ultra-low-noise-platinum-atx-1500-watt-pc-power-supply-cp-9020309-na) | Corsair HX1500i$320 · [Corsair store price](https://www.corsair.com/us/en/p/psu/cp-9020309-na/hx1500i-fully-modular-ultra-low-noise-platinum-atx-1500-watt-pc-power-supply-cp-9020309-na) | Corsair HX1200i$280 · [Corsair store price](https://www.corsair.com/us/en/p/psu/cp-9020307-na/hx1200i-fully-modular-ultra-low-noise-platinum-atx-1200-watt-pc-power-supply-cp-9020307-na) | | Case | Corsair FRAME 5000D RS$160 · [Corsair store price](https://www.corsair.com/us/en/p/pc-cases/cc-9011307-ww/frame-5000d-rs-high-airflow-mid-tower-pc-case-cc-9011307-ww) | Corsair FRAME 5000D RS$160 · [Corsair store price](https://www.corsair.com/us/en/p/pc-cases/cc-9011307-ww/frame-5000d-rs-high-airflow-mid-tower-pc-case-cc-9011307-ww) | Corsair FRAME 5000D RS$160 · [Corsair store price](https://www.corsair.com/us/en/p/pc-cases/cc-9011307-ww/frame-5000d-rs-high-airflow-mid-tower-pc-case-cc-9011307-ww) | | Storage | Samsung 990 2TB NVMe$330 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16820147923) | Samsung 990 2TB NVMe$330 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16820147923) | Samsung 990 2TB NVMe$330 · [Newegg price (sold by Newegg)](https://www.newegg.com/p/N82E16820147923) | We may earn a commission when you buy through these links, at no extra cost to you. Prices are dated snapshots (2026-10-23 for parts; GPUs at list price (MSRP) unless marked used; used prices are one dated listing each). ## Fit checks ### A: Dual used RTX 3090 build (48 GB) for 32B to 70B models * PSU: 1500 W vs. 1326 W recommended ((GPU 700 W + CPU 170 W + 150 W) × 1.3). Headroom 174 W. * Power connectors: need 4× PCIe 8-pin; PSU has 2× 12V-2x6 + 5× 8-pin. * Length: 313 mm card vs. 450 mm case limit. * Expansion slots: 2 cards × 3-slot needs about 6; case has 8. * Motherboard: 2 GPUs on 2 x16-size slots (2x PCIe 5.0 x16 (CPU; x16 or x8/x8)). * Slot spacing: check two 3-slot cards physically fit side by side on this board. We don’t model slot positions. ### B: Dual RX 7900 XTX build (48 GB VRAM) * PSU: 1500 W vs. 1339 W recommended ((GPU 710 W + CPU 170 W + 150 W) × 1.3). Headroom 161 W. * Power connectors: need 4× PCIe 8-pin; PSU has 2× 12V-2x6 + 5× 8-pin. * Length: 287 mm card vs. 450 mm case limit. * Expansion slots: 2 cards × 2.5-slot needs about 6; case has 8. * Motherboard: 2 GPUs on 2 x16-size slots (2x PCIe 5.0 x16 (CPU; x16 or x8/x8)). * Slot spacing: check two 2.5-slot cards physically fit side by side on this board. We don’t model slot positions. ### C: RTX 5090 single-GPU build (32 GB) * PSU: 1200 W vs. 1027 W recommended ((GPU 575 W + CPU 65 W + 150 W) × 1.3). Headroom 173 W. * Power connectors: need 1× 12V-2x6 cable + 0× PCIe 8-pin; PSU has 2× 12V-2x6 + 5× 8-pin. * Length: 304 mm card vs. 450 mm case limit. * Expansion slots: 1 card × 2-slot needs about 2; case has 8. ## Every model on each build At the quant above, about 4k context. Max context isn't modeled yet (no sourced KV-cache data). __Every tracked model at Q4 on each build__ | Model | A: Dual used RTX 3090 build (48 GB) for 32B to 70B models | B: Dual RX 7900 XTX build (48 GB VRAM) | C: RTX 5090 single-GPU build (32 GB) | | ---------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------- | -------------------------------------- | ------------------------------------ | | [Kimi-K3.1-Pro](https://whatcanirunai.vercel.app/models/kimi-k3.1-pro/)1939.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K3](https://whatcanirunai.vercel.app/models/kimi-k3/)1939.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3.8-2.4T-A95B](https://whatcanirunai.vercel.app/models/qwen3.8-2.4t-a95b/)1706.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [LongCat-2.5](https://whatcanirunai.vercel.app/models/longcat-2.5/)1241.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [LongCat-2.0](https://whatcanirunai.vercel.app/models/longcat-2.0/)1238.9 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V4-Pro-0813](https://whatcanirunai.vercel.app/models/deepseek-v4-pro-0813/)1151.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V4-Pro](https://whatcanirunai.vercel.app/models/deepseek-v4-pro/)1115.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2.5](https://whatcanirunai.vercel.app/models/kimi-k2.5/)716.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2.6](https://whatcanirunai.vercel.app/models/kimi-k2.6/)716.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2.7-Code](https://whatcanirunai.vercel.app/models/kimi-k2.7-code/)716.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2-Thinking](https://whatcanirunai.vercel.app/models/kimi-k2-thinking/)716.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2-Instruct-0905](https://whatcanirunai.vercel.app/models/kimi-k2-instruct-0905/)716.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kimi-K2-Instruct](https://whatcanirunai.vercel.app/models/kimi-k2-instruct/)716.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.6-Pro-RL](https://whatcanirunai.vercel.app/models/mimo-v2.6-pro-rl/)715.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.6-Pro-MOPD](https://whatcanirunai.vercel.app/models/mimo-v2.6-pro-mopd/)715.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.5-Pro](https://whatcanirunai.vercel.app/models/mimo-v2.5-pro/)714.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ring-2.5-1T](https://whatcanirunai.vercel.app/models/ring-2.5-1t/)706.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-2.5-1T](https://whatcanirunai.vercel.app/models/ling-2.5-1t/)706.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-1T](https://whatcanirunai.vercel.app/models/ling-1t/)698.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ring-1T](https://whatcanirunai.vercel.app/models/ring-1t/)698.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Inkling](https://whatcanirunai.vercel.app/models/inkling/)665.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Intern-S1-Pro](https://whatcanirunai.vercel.app/models/intern-s1-pro/)640.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Muse-Spark-Open-816B-A42B](https://whatcanirunai.vercel.app/models/muse-spark-open-816b-a42b/)574.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Hy4-preview](https://whatcanirunai.vercel.app/models/hy4-preview/)544.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-5](https://whatcanirunai.vercel.app/models/glm-5/)526.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-5.1](https://whatcanirunai.vercel.app/models/glm-5.1/)526.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-5.3](https://whatcanirunai.vercel.app/models/glm-5.3/)526.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [GLM-5.2](https://whatcanirunai.vercel.app/models/glm-5.2/)526.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-5.3-BF16](https://whatcanirunai.vercel.app/models/glm-5.3-bf16/)526.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [K-EXAONE-2.0-750B-A37B](https://whatcanirunai.vercel.app/models/k-exaone-2.0-750b-a37b/)523.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V3.2-Exp](https://whatcanirunai.vercel.app/models/deepseek-v3.2-exp/)478.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V3.2-Speciale](https://whatcanirunai.vercel.app/models/deepseek-v3.2-speciale/)478.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V3.2](https://whatcanirunai.vercel.app/models/deepseek-v3.2/)478.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V3-0324](https://whatcanirunai.vercel.app/models/deepseek-v3-0324/)478.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-R1-0528](https://whatcanirunai.vercel.app/models/deepseek-r1-0528/)478.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V3.1-Base](https://whatcanirunai.vercel.app/models/deepseek-v3.1-base/)478.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-R1](https://whatcanirunai.vercel.app/models/deepseek-r1/)478.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-R1-Zero](https://whatcanirunai.vercel.app/models/deepseek-r1-zero/)478.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Step-5-Preview](https://whatcanirunai.vercel.app/models/step-5-preview/)422.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.1-flash](https://whatcanirunai.vercel.app/models/ling-3.1-flash/)399.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.1-flash-1019](https://whatcanirunai.vercel.app/models/ling-3.1-flash-1019/)399.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [LongCat-Flash-Chat](https://whatcanirunai.vercel.app/models/longcat-flash-chat/)392.7 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [LongCat-Flash-Thinking](https://whatcanirunai.vercel.app/models/longcat-flash-thinking/)392.7 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [LongCat-Flash-Thinking-2601](https://whatcanirunai.vercel.app/models/longcat-flash-thinking-2601/)392.7 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16](https://whatcanirunai.vercel.app/models/nvidia-nemotron-3-ultra-550b-a55b-bf16/)391.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Beam-1.0-501B-A23B](https://whatcanirunai.vercel.app/models/beam-1.0-501b-a23b/)350.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-Coder-480B-A35B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-coder-480b-a35b-instruct/)335.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M1-40k](https://whatcanirunai.vercel.app/models/minimax-m1-40k/)319.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M1-80k](https://whatcanirunai.vercel.app/models/minimax-m1-80k/)319.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M3](https://whatcanirunai.vercel.app/models/minimax-m3/)298.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [ERNIE-4.5-VL-424B-A47B-PT](https://whatcanirunai.vercel.app/models/ernie-4.5-vl-424b-a47b-pt/)296.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Intern-S2-397B](https://whatcanirunai.vercel.app/models/intern-s2-397b/)282.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3.5-397B-A17B](https://whatcanirunai.vercel.app/models/qwen3.5-397b-a17b/)282.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Llama-4-Maverick-17B-128E-Instruct](https://whatcanirunai.vercel.app/models/llama-4-maverick-17b-128e-instruct/)281.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Trinity-Large-Thinking](https://whatcanirunai.vercel.app/models/trinity-large-thinking/)278.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Trinity-Large-Preview](https://whatcanirunai.vercel.app/models/trinity-large-preview/)278.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Trinity-Large-TrueBase](https://whatcanirunai.vercel.app/models/trinity-large-truebase/)278.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Trinity-Large-Base](https://whatcanirunai.vercel.app/models/trinity-large-base/)278.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Nex-N2.5-Pro](https://whatcanirunai.vercel.app/models/nex-n2.5-pro/)277.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Nex-N2-Pro](https://whatcanirunai.vercel.app/models/nex-n2-pro/)277.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [NorthCUA-Pro](https://whatcanirunai.vercel.app/models/northcua-pro/)277.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.5](https://whatcanirunai.vercel.app/models/glm-4.5/)250.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.7](https://whatcanirunai.vercel.app/models/glm-4.7/)250.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.6](https://whatcanirunai.vercel.app/models/glm-4.6/)249.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-5.3-Flash](https://whatcanirunai.vercel.app/models/glm-5.3-flash/)225.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Motif-3](https://whatcanirunai.vercel.app/models/motif-3/)220.5 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.5](https://whatcanirunai.vercel.app/models/mimo-v2.5/)217.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.6-Flash-RL](https://whatcanirunai.vercel.app/models/mimo-v2.6-flash-rl/)217.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2.6-Flash-MOPD](https://whatcanirunai.vercel.app/models/mimo-v2.6-flash-mopd/)217.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiMo-V2-Flash](https://whatcanirunai.vercel.app/models/mimo-v2-flash/)217.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V4-Flash-Vision-Exp](https://whatcanirunai.vercel.app/models/deepseek-v4-flash-vision-exp/)213.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V4-Flash-0731](https://whatcanirunai.vercel.app/models/deepseek-v4-flash-0731/)213.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [ERNIE-4.5-300B-A47B-PT](https://whatcanirunai.vercel.app/models/ernie-4.5-300b-a47b-pt/)210.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Hy3](https://whatcanirunai.vercel.app/models/hy3/)209.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Hy3-preview](https://whatcanirunai.vercel.app/models/hy3-preview/)209.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-plus](https://whatcanirunai.vercel.app/models/ling-plus/)205.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V4-Flash](https://whatcanirunai.vercel.app/models/deepseek-v4-flash/)203.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MemT-1.0-Flash](https://whatcanirunai.vercel.app/models/memt-1.0-flash/)201.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [ERNIE-5.5-Flash-Base](https://whatcanirunai.vercel.app/models/ernie-5.5-flash-base/)200.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Inkling-Small](https://whatcanirunai.vercel.app/models/inkling-small/)186.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Llama-3\_1-Nemotron-Ultra-253B-v1](https://whatcanirunai.vercel.app/models/llama-3-1-nemotron-ultra-253b-v1/)177.7 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Solar-Open2-250B](https://whatcanirunai.vercel.app/models/solar-open2-250b/)175.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Intern-S1](https://whatcanirunai.vercel.app/models/intern-s1/)168.8 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [K-EXAONE-236B-A23B](https://whatcanirunai.vercel.app/models/k-exaone-236b-a23b/)166.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V2.5](https://whatcanirunai.vercel.app/models/deepseek-v2.5/)165.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-Coder-V2-Instruct](https://whatcanirunai.vercel.app/models/deepseek-coder-v2-instruct/)165.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V2.5-1210](https://whatcanirunai.vercel.app/models/deepseek-v2.5-1210/)165.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-V2-Chat-0628](https://whatcanirunai.vercel.app/models/deepseek-v2-chat-0628/)165.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-Coder-V2-Instruct-0724](https://whatcanirunai.vercel.app/models/deepseek-coder-v2-instruct-0724/)165.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-VL-235B-A22B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-235b-a22b-instruct/)165.3 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-235B-A22B](https://whatcanirunai.vercel.app/models/qwen3-235b-a22b/)164.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-235B-A22B-Instruct-2507](https://whatcanirunai.vercel.app/models/qwen3-235b-a22b-instruct-2507/)164.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-235B-A22B-Thinking-2507](https://whatcanirunai.vercel.app/models/qwen3-235b-a22b-thinking-2507/)164.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M2](https://whatcanirunai.vercel.app/models/minimax-m2/)160.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M2.5](https://whatcanirunai.vercel.app/models/minimax-m2.5/)160.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M2.7](https://whatcanirunai.vercel.app/models/minimax-m2.7/)160.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [MiniMax-M2.1](https://whatcanirunai.vercel.app/models/minimax-m2.1/)160.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [command-a-plus-05-2026-bf16](https://whatcanirunai.vercel.app/models/command-a-plus-05-2026-bf16/)153.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Step-3.7-Flash](https://whatcanirunai.vercel.app/models/step-3.7-flash/)141.4 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Step-3.5-Flash](https://whatcanirunai.vercel.app/models/step-3.5-flash/)140.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Step-3.5-Flash-Base-Midtrain](https://whatcanirunai.vercel.app/models/step-3.5-flash-base-midtrain/)138.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3.8-Flash-Next](https://whatcanirunai.vercel.app/models/qwen3.8-flash-next/)126.5 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Mixtral-8x22B-Instruct-v0.1](https://whatcanirunai.vercel.app/models/mixtral-8x22b-instruct-v0.1/)99.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.0-flash-Sante](https://whatcanirunai.vercel.app/models/ling-3.0-flash-sante/)89.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.0-flash](https://whatcanirunai.vercel.app/models/ling-3.0-flash/)89.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.0-flash-Fin](https://whatcanirunai.vercel.app/models/ling-3.0-flash-fin/)89.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ornith-1.5-122B](https://whatcanirunai.vercel.app/models/ornith-1.5-122b/)88.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Qwen3.5-122B-A10B](https://whatcanirunai.vercel.app/models/qwen3.5-122b-a10b/)88.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Devstral-2-123B-Instruct-2512](https://whatcanirunai.vercel.app/models/devstral-2-123b-instruct-2512/)88.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Ling-3.0-flash-VL](https://whatcanirunai.vercel.app/models/ling-3.0-flash-vl/)88.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [NVIDIA-Nemotron-3-Super-120B-A12B-BF16](https://whatcanirunai.vercel.app/models/nvidia-nemotron-3-super-120b-a12b-bf16/)87.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Mistral-Small-4-119B-2603](https://whatcanirunai.vercel.app/models/mistral-small-4-119b-2603/)84.2 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Youtu-LLM-A7B-Base](https://whatcanirunai.vercel.app/models/youtu-llm-a7b-base/)83.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Laguna-S-2.1](https://whatcanirunai.vercel.app/models/laguna-s-2.1/)83.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [gpt-oss-120b](https://whatcanirunai.vercel.app/models/gpt-oss-120b/)82.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [c4ai-command-a-03-2025](https://whatcanirunai.vercel.app/models/c4ai-command-a-03-2025/)78.4 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.5-Air](https://whatcanirunai.vercel.app/models/glm-4.5-air/)78.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Llama-4-Scout-17B-16E-Instruct](https://whatcanirunai.vercel.app/models/llama-4-scout-17b-16e-instruct/)76.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Llama-4-Scout-17B-16E](https://whatcanirunai.vercel.app/models/llama-4-scout-17b-16e/)76.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.5V](https://whatcanirunai.vercel.app/models/glm-4.5v/)76.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [GLM-4.6V](https://whatcanirunai.vercel.app/models/glm-4.6v/)76.1 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-2.6-flash](https://whatcanirunai.vercel.app/models/ling-2.6-flash/)75.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [sarvam-105b](https://whatcanirunai.vercel.app/models/sarvam-105b/)74.9 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ling-flash-2.0](https://whatcanirunai.vercel.app/models/ling-flash-2.0/)72.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [LLaDA2.1-flash](https://whatcanirunai.vercel.app/models/llada2.1-flash/)72.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Ring-flash-2.0](https://whatcanirunai.vercel.app/models/ring-flash-2.0/)72.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [LLaDA2.2-flash](https://whatcanirunai.vercel.app/models/llada2.2-flash/)72.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Solar-Open-100B](https://whatcanirunai.vercel.app/models/solar-open-100b/)72.6 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-Next-80B-A3B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-next-80b-a3b-instruct/)57.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-Next-80B-A3B-Thinking](https://whatcanirunai.vercel.app/models/qwen3-next-80b-a3b-thinking/)57.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [AliceAI-Foundation-80B-A3B-Base](https://whatcanirunai.vercel.app/models/aliceai-foundation-80b-a3b-base/)57.7 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Hunyuan-A13B-Instruct](https://whatcanirunai.vercel.app/models/hunyuan-a13b-instruct/)57.0 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Qwen3-Coder-Next](https://whatcanirunai.vercel.app/models/qwen3-coder-next/)56.5 GB at Q4 · MoE | Won’t fit | Won’t fit | Won’t fit | | [Kolibri-1](https://whatcanirunai.vercel.app/models/kolibri-1/)55.5 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Qwen2.5-VL-72B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-vl-72b-instruct/)52.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [UI-TARS-72B-DPO](https://whatcanirunai.vercel.app/models/ui-tars-72b-dpo/)52.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Qwen2.5-72B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-72b-instruct/)51.7 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Apertus-70B-Instruct-2509](https://whatcanirunai.vercel.app/models/apertus-70b-instruct-2509/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Apertus-70B-2509](https://whatcanirunai.vercel.app/models/apertus-70b-2509/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Llama-3.3-70B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.3-70b-instruct/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [DeepSeek-R1-Distill-Llama-70B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-llama-70b/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Llama-3.1-70B](https://whatcanirunai.vercel.app/models/llama-3.1-70b/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Hermes-2-Theta-Llama-3-70B](https://whatcanirunai.vercel.app/models/hermes-2-theta-llama-3-70b/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [Meta-Llama-3.1-70B-Instruct](https://whatcanirunai.vercel.app/models/meta-llama-3.1-70b-instruct/)50.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [LongCat-Flash-Lite](https://whatcanirunai.vercel.app/models/longcat-flash-lite/)49.2 GB at Q4 | Won’t fit | Won’t fit | Won’t fit | | [AI21-Jamba2-Mini](https://whatcanirunai.vercel.app/models/ai21-jamba2-mini/)37.0 GB at Q4 · MoE | Fitsup to 109 tok/s | Fitsup to 112 tok/s | Won’t fit | | [Llama-3\_1-Nemotron-51B-Instruct](https://whatcanirunai.vercel.app/models/llama-3-1-nemotron-51b-instruct/)36.9 GB at Q4 | Fits18–25 tok/s | Fits18–26 tok/s | Won’t fit | | [Llama-3\_3-Nemotron-Super-49B-v1\_5](https://whatcanirunai.vercel.app/models/llama-3-3-nemotron-super-49b-v1-5/)35.8 GB at Q4 | Fits19–26 tok/s | Fits19–27 tok/s | Won’t fit | | [Llama-3\_3-Nemotron-Super-49B-v1](https://whatcanirunai.vercel.app/models/llama-3-3-nemotron-super-49b-v1/)35.8 GB at Q4 | Fits19–26 tok/s | Fits19–27 tok/s | Won’t fit | | [Phi-3.5-MoE-instruct](https://whatcanirunai.vercel.app/models/phi-3.5-moe-instruct/)30.2 GB at Q4 · MoE | Fitsup to 199 tok/s | Fitsup to 204 tok/s | Tightup to 381 tok/s | | [GRIN-MoE](https://whatcanirunai.vercel.app/models/grin-moe/)30.2 GB at Q4 · MoE | Fitsup to 199 tok/s | Fitsup to 204 tok/s | Tightup to 381 tok/s | | [ALIA-40b](https://whatcanirunai.vercel.app/models/alia-40b/)29.2 GB at Q4 | Fits23–32 tok/s | Fits23–33 tok/s | Tight44–62 tok/s | | [IQuest-Coder-V1-40B-Loop-Instruct](https://whatcanirunai.vercel.app/models/iquest-coder-v1-40b-loop-instruct/)28.7 GB at Q4 | Fits23–33 tok/s | Fits24–34 tok/s | Fits45–63 tok/s | | [IQuest-Coder-V1-40B-Instruct](https://whatcanirunai.vercel.app/models/iquest-coder-v1-40b-instruct/)28.7 GB at Q4 | Fits23–33 tok/s | Fits24–34 tok/s | Fits45–63 tok/s | | [Recursive Flint](https://whatcanirunai.vercel.app/models/flint/)28.7 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 539 tok/s | Fitsup to 1006 tok/s | | [K2-Horizon-MoVA-36B-A4B](https://whatcanirunai.vercel.app/models/k2-horizon-mova-36b-a4b/)27.1 GB at Q4 · MoE | Fitsup to 328 tok/s | Fitsup to 337 tok/s | Fitsup to 628 tok/s | | [Seed-OSS-36B-Instruct](https://whatcanirunai.vercel.app/models/seed-oss-36b-instruct/)26.2 GB at Q4 | Fits26–36 tok/s | Fits26–37 tok/s | Fits49–69 tok/s | | [Seed-OSS-36B-Base-woSyn](https://whatcanirunai.vercel.app/models/seed-oss-36b-base-wosyn/)26.2 GB at Q4 | Fits26–36 tok/s | Fits26–37 tok/s | Fits49–69 tok/s | | [Intern-S2-Preview](https://whatcanirunai.vercel.app/models/intern-s2-preview/)26.2 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Ornith-1.5-35B-A3B](https://whatcanirunai.vercel.app/models/ornith-1.5-35b-a3b/)26.1 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Sage-A1.5](https://whatcanirunai.vercel.app/models/sage-a1.5/)26.1 GB at Q4 | Fits26–37 tok/s | Fits26–37 tok/s | Fits49–70 tok/s | | [Qwen3.6-35B-A3B](https://whatcanirunai.vercel.app/models/qwen3.6-35b-a3b/)26.1 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Qwen3.5-35B-A3B](https://whatcanirunai.vercel.app/models/qwen3.5-35b-a3b/)26.1 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Apex-35B-V1.0](https://whatcanirunai.vercel.app/models/apex-35b-v1.0/)26.1 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [MiniCPM-V-4.7-35B-A3B](https://whatcanirunai.vercel.app/models/minicpm-v-4.7-35b-a3b/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [MemT-1.0-Nano](https://whatcanirunai.vercel.app/models/memt-1.0-nano/)25.5 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Agents-A1](https://whatcanirunai.vercel.app/models/agents-a1/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Nex-N2.5-mini](https://whatcanirunai.vercel.app/models/nex-n2.5-mini/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Nex-N2-mini](https://whatcanirunai.vercel.app/models/nex-n2-mini/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Axion-1.0-35B](https://whatcanirunai.vercel.app/models/axion-1.0-35b/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [NorthCUA-Mini](https://whatcanirunai.vercel.app/models/northcua-mini/)25.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Clover-35B-A3B-Preview](https://whatcanirunai.vercel.app/models/clover-35b-a3b-preview/)25.2 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen-AgentWorld-35B-A3B](https://whatcanirunai.vercel.app/models/qwen-agentworld-35b-a3b/)25.2 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [KAT-Coder-V2.5-Dev](https://whatcanirunai.vercel.app/models/kat-coder-v2.5-dev/)25.2 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [EXAONE-4.5-33B](https://whatcanirunai.vercel.app/models/exaone-4.5-33b/)24.9 GB at Q4 | Fits27–38 tok/s | Fits28–39 tok/s | Fits52–73 tok/s | | [Falcon-H1-34B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-34b-instruct/)24.5 GB at Q4 | Fits28–39 tok/s | Fits28–40 tok/s | Fits53–75 tok/s | | [Qwen2.5-VL-32B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-vl-32b-instruct/)24.3 GB at Q4 | Fits28–39 tok/s | Fits28–40 tok/s | Fits53–75 tok/s | | [Laguna-XS.2](https://whatcanirunai.vercel.app/models/laguna-xs.2/)24.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Qwen3-VL-32B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-32b-instruct/)24.3 GB at Q4 | Fits28–39 tok/s | Fits28–40 tok/s | Fits53–75 tok/s | | [HyperCLOVAX-SEED-Think-32B](https://whatcanirunai.vercel.app/models/hyperclovax-seed-think-32b/)24.2 GB at Q4 | Fits28–39 tok/s | Fits29–40 tok/s | Fits53–75 tok/s | | [FlexOlmo-7x7B-1T](https://whatcanirunai.vercel.app/models/flexolmo-7x7b-1t/)24.2 GB at Q4 | Fits28–39 tok/s | Fits29–40 tok/s | Fits53–76 tok/s | | [GLM-Z1-Rumination-32B-0414](https://whatcanirunai.vercel.app/models/glm-z1-rumination-32b-0414/)24.1 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–76 tok/s | | [Nanbeige4.3-33B](https://whatcanirunai.vercel.app/models/nanbeige4.3-33b/)24.1 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–76 tok/s | | [Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16](https://whatcanirunai.vercel.app/models/nemotron-3-nano-omni-30b-a3b-reasoning-bf16/)24.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen2.5-Coder-32B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-32b-instruct/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [DeepSeek-R1-Distill-Qwen-32B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-qwen-32b/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [Qwen2.5-32B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-32b-instruct/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [QwQ-32B](https://whatcanirunai.vercel.app/models/qwq-32b/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [QwQ-32B-Preview](https://whatcanirunai.vercel.app/models/qwq-32b-preview/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [Qwen2.5-32B](https://whatcanirunai.vercel.app/models/qwen2.5-32b/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [Qwen2.5-Coder-32B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-32b/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [Qwen3-32B](https://whatcanirunai.vercel.app/models/qwen3-32b/)23.8 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [GLM-Z1-32B-0414](https://whatcanirunai.vercel.app/models/glm-z1-32b-0414/)23.7 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [GLM-4-32B-0414](https://whatcanirunai.vercel.app/models/glm-4-32b-0414/)23.7 GB at Q4 | Fits28–40 tok/s | Fits29–41 tok/s | Fits54–77 tok/s | | [OLMo-2-0325-32B-Instruct](https://whatcanirunai.vercel.app/models/olmo-2-0325-32b-instruct/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [OLMo-2-0325-32B](https://whatcanirunai.vercel.app/models/olmo-2-0325-32b/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [Olmo-3-32B-Think](https://whatcanirunai.vercel.app/models/olmo-3-32b-think/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [Olmo-3-1125-32B](https://whatcanirunai.vercel.app/models/olmo-3-1125-32b/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [Olmo-3.1-32B-Instruct](https://whatcanirunai.vercel.app/models/olmo-3.1-32b-instruct/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [Olmo-3.1-32B-Think](https://whatcanirunai.vercel.app/models/olmo-3.1-32b-think/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [Olmo-3.1-32B-Instruct-DPO](https://whatcanirunai.vercel.app/models/olmo-3.1-32b-instruct-dpo/)23.5 GB at Q4 | Fits29–41 tok/s | Fits29–42 tok/s | Fits55–78 tok/s | | [granite-4.0-h-small](https://whatcanirunai.vercel.app/models/granite-4.0-h-small/)23.5 GB at Q4 · MoE | Fitsup to 146 tok/s | Fitsup to 150 tok/s | Fitsup to 279 tok/s | | [sarvam-30b](https://whatcanirunai.vercel.app/models/sarvam-30b/)23.4 GB at Q4 · MoE | Fitsup to 547 tok/s | Fitsup to 561 tok/s | Fitsup to 1047 tok/s | | [EXAONE-4.0-32B](https://whatcanirunai.vercel.app/models/exaone-4.0-32b/)23.3 GB at Q4 | Fits29–41 tok/s | Fits30–42 tok/s | Fits55–79 tok/s | | [EXAONE-4.0.1-32B](https://whatcanirunai.vercel.app/models/exaone-4.0.1-32b/)23.3 GB at Q4 | Fits29–41 tok/s | Fits30–42 tok/s | Fits55–79 tok/s | | [EXAONE-3.5-32B-Instruct](https://whatcanirunai.vercel.app/models/exaone-3.5-32b-instruct/)23.3 GB at Q4 | Fits29–41 tok/s | Fits30–42 tok/s | Fits55–79 tok/s | | [EXAONE-Deep-32B](https://whatcanirunai.vercel.app/models/exaone-deep-32b/)23.3 GB at Q4 | Fits29–41 tok/s | Fits30–42 tok/s | Fits55–79 tok/s | | [NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16](https://whatcanirunai.vercel.app/models/nvidia-nemotron-3.5-lightning-30b-a3b-bf16/)23.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [NVIDIA-Nemotron-3-Nano-30B-A3B-BF16](https://whatcanirunai.vercel.app/models/nvidia-nemotron-3-nano-30b-a3b-bf16/)23.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Nemotron-Cascade-2-30B-A3B](https://whatcanirunai.vercel.app/models/nemotron-cascade-2-30b-a3b/)23.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Gemma 4 31B](https://whatcanirunai.vercel.app/models/gemma-4-31b-it/)22.8 GB at Q4 | Fits30–42 tok/s | Fits30–43 tok/s | Fits57–80 tok/s | | [gemma-4-31B](https://whatcanirunai.vercel.app/models/gemma-4-31b/)22.8 GB at Q4 | Fits30–42 tok/s | Fits30–43 tok/s | Fits57–80 tok/s | | [GLM-4.7-Flash](https://whatcanirunai.vercel.app/models/glm-4.7-flash/)22.8 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [Xing4.0-29B-A4B](https://whatcanirunai.vercel.app/models/xing4.0-29b-a4b/)22.8 GB at Q4 · MoE | Fitsup to 328 tok/s | Fitsup to 337 tok/s | Fitsup to 628 tok/s | | [Xing4.0-29B-Intent](https://whatcanirunai.vercel.app/models/xing4.0-29b-intent/)22.8 GB at Q4 · MoE | Fitsup to 328 tok/s | Fitsup to 337 tok/s | Fitsup to 628 tok/s | | [Qwen3-VL-30B-A3B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-30b-a3b-instruct/)22.7 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen3-VL-30B-A3B-Thinking](https://whatcanirunai.vercel.app/models/qwen3-vl-30b-a3b-thinking/)22.7 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [UI-Venus-1.5-30B-A3B](https://whatcanirunai.vercel.app/models/ui-venus-1.5-30b-a3b/)22.7 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [kanana-2-30b-a3b-instruct-2601](https://whatcanirunai.vercel.app/models/kanana-2-30b-a3b-instruct-2601/)22.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [kanana-2-30b-a3b-instruct](https://whatcanirunai.vercel.app/models/kanana-2-30b-a3b-instruct/)22.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [kanana-2-30b-a3b-thinking-2601](https://whatcanirunai.vercel.app/models/kanana-2-30b-a3b-thinking-2601/)22.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Qwen3-30B-A3B](https://whatcanirunai.vercel.app/models/qwen3-30b-a3b/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen3-Coder-30B-A3B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-coder-30b-a3b-instruct/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen3-30B-A3B-Instruct-2507](https://whatcanirunai.vercel.app/models/qwen3-30b-a3b-instruct-2507/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Qwen3-30B-A3B-Thinking-2507](https://whatcanirunai.vercel.app/models/qwen3-30b-a3b-thinking-2507/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Tongyi-DeepResearch-30B-A3B](https://whatcanirunai.vercel.app/models/tongyi-deepresearch-30b-a3b/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [Rubicon-Preview](https://whatcanirunai.vercel.app/models/rubicon-preview/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 448 tok/s | Fitsup to 837 tok/s | | [North-Mini-Code-1.0](https://whatcanirunai.vercel.app/models/north-mini-code-1.0/)22.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Hy-Memory-30B-A3B](https://whatcanirunai.vercel.app/models/hy-memory-30b-a3b/)22.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [Hy-MT2-30B-A3B](https://whatcanirunai.vercel.app/models/hy-mt2-30b-a3b/)22.0 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [granite-switch-4.1-30b](https://whatcanirunai.vercel.app/models/granite-switch-4.1-30b/)21.9 GB at Q4 | Fits31–44 tok/s | Fits32–45 tok/s | Fits59–84 tok/s | | [Muse-Glimmer-30B](https://whatcanirunai.vercel.app/models/muse-glimmer-30b/)21.8 GB at Q4 | Fits31–44 tok/s | Fits32–45 tok/s | Fits60–84 tok/s | | [ERNIE-4.5-VL-28B-A3B-Thinking](https://whatcanirunai.vercel.app/models/ernie-4.5-vl-28b-a3b-thinking/)21.7 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [ERNIE-4.5-VL-28B-A3B-PT](https://whatcanirunai.vercel.app/models/ernie-4.5-vl-28b-a3b-pt/)21.5 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [granite-4.2-30b](https://whatcanirunai.vercel.app/models/granite-4.2-30b/)21.4 GB at Q4 | Fits32–45 tok/s | Fits32–46 tok/s | Fits61–86 tok/s | | [granite-4.1-30b](https://whatcanirunai.vercel.app/models/granite-4.1-30b/)21.1 GB at Q4 | Fits32–45 tok/s | Fits33–47 tok/s | Fits61–87 tok/s | | [Raon-1 mini preview](https://whatcanirunai.vercel.app/models/raon-1-mini-preview/)21.1 GB at Q4 · MoE | Fitsup to 305 tok/s | Fitsup to 313 tok/s | Fitsup to 584 tok/s | | [Qwen3.8-27B](https://whatcanirunai.vercel.app/models/qwen3.8-27b/)20.4 GB at Q4 | Fits33–47 tok/s | Fits34–48 tok/s | Fits64–90 tok/s | | [Tess-4.5-27B](https://whatcanirunai.vercel.app/models/tess-4.5-27b/)20.4 GB at Q4 | Fits33–47 tok/s | Fits34–48 tok/s | Fits64–90 tok/s | | [Qwen3.6-27B](https://whatcanirunai.vercel.app/models/qwen3.6-27b/)20.4 GB at Q4 | Fits33–47 tok/s | Fits34–48 tok/s | Fits64–90 tok/s | | [Qwen3.5-27B](https://whatcanirunai.vercel.app/models/qwen3.5-27b/)20.4 GB at Q4 | Fits33–47 tok/s | Fits34–48 tok/s | Fits64–90 tok/s | | [gemma-3-27b-it](https://whatcanirunai.vercel.app/models/gemma-3-27b-it/)20.1 GB at Q4 | Fits34–48 tok/s | Fits35–49 tok/s | Fits65–92 tok/s | | [Arete-27B](https://whatcanirunai.vercel.app/models/arete-27b/)20.1 GB at Q4 | Fits34–48 tok/s | Fits35–49 tok/s | Fits65–92 tok/s | | [gemma-2-27b-it](https://whatcanirunai.vercel.app/models/gemma-2-27b-it/)20.0 GB at Q4 | Fits34–48 tok/s | Fits35–49 tok/s | Fits65–92 tok/s | | [gemma-2-27b](https://whatcanirunai.vercel.app/models/gemma-2-27b/)20.0 GB at Q4 | Fits34–48 tok/s | Fits35–49 tok/s | Fits65–92 tok/s | | [Qwen3.8-27B-Longbook27-v2](https://whatcanirunai.vercel.app/models/qwen3.8-27b-longbook27-v2/)19.9 GB at Q4 | Fits34–48 tok/s | Fits35–50 tok/s | Fits66–93 tok/s | | [medgemma-27b-text-it](https://whatcanirunai.vercel.app/models/medgemma-27b-text-it/)19.8 GB at Q4 | Fits34–49 tok/s | Fits35–50 tok/s | Fits66–93 tok/s | | [Hemmingway-1.5](https://whatcanirunai.vercel.app/models/hemmingway-1.5/)19.8 GB at Q4 | Fits34–49 tok/s | Fits35–50 tok/s | Fits66–93 tok/s | | [Trinity-Mini](https://whatcanirunai.vercel.app/models/trinity-mini/)19.2 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [diffusiongemma-26B-A4B-it](https://whatcanirunai.vercel.app/models/diffusiongemma-26b-a4b-it/)19.0 GB at Q4 · MoE | Fitsup to 328 tok/s | Fitsup to 336 tok/s | Fitsup to 628 tok/s | | [Gemma 4 26B A4B](https://whatcanirunai.vercel.app/models/gemma-4-26b-a4b-it/)19.0 GB at Q4 · MoE | Fitsup to 345 tok/s | Fitsup to 354 tok/s | Fitsup to 661 tok/s | | [gemma-4-26B-A4B](https://whatcanirunai.vercel.app/models/gemma-4-26b-a4b/)19.0 GB at Q4 · MoE | Fitsup to 328 tok/s | Fitsup to 337 tok/s | Fitsup to 628 tok/s | | [Voxtral-Small-24B-2507](https://whatcanirunai.vercel.app/models/voxtral-small-24b-2507/)17.9 GB at Q4 | Fits38–54 tok/s | Fits39–55 tok/s | Fits73–104 tok/s | | [Mistral-Small-3.2-24B-Instruct-2506](https://whatcanirunai.vercel.app/models/mistral-small-3.2-24b-instruct-2506/)17.7 GB at Q4 | Fits39–55 tok/s | Fits40–56 tok/s | Fits74–105 tok/s | | [Mistral-Small-3.1-24B-Instruct-2503](https://whatcanirunai.vercel.app/models/mistral-small-3.1-24b-instruct-2503/)17.7 GB at Q4 | Fits39–55 tok/s | Fits40–56 tok/s | Fits74–105 tok/s | | [LFM2.5-24B-A2B](https://whatcanirunai.vercel.app/models/lfm2.5-24b-a2b/)17.7 GB at Q4 · MoE | Fitsup to 656 tok/s | Fitsup to 673 tok/s | Fitsup to 1257 tok/s | | [LFM2-24B-A2B](https://whatcanirunai.vercel.app/models/lfm2-24b-a2b/)17.6 GB at Q4 · MoE | Fitsup to 656 tok/s | Fitsup to 673 tok/s | Fitsup to 1256 tok/s | | [Mistral-Small-24B-Instruct-2501](https://whatcanirunai.vercel.app/models/mistral-small-24b-instruct-2501/)17.4 GB at Q4 | Fits39–56 tok/s | Fits40–57 tok/s | Fits75–107 tok/s | | [EuroLLM-22B-Instruct-2512](https://whatcanirunai.vercel.app/models/eurollm-22b-instruct-2512/)16.8 GB at Q4 | Fits41–58 tok/s | Fits42–59 tok/s | Fits78–111 tok/s | | [EuroLLM-22B-2512](https://whatcanirunai.vercel.app/models/eurollm-22b-2512/)16.8 GB at Q4 | Fits41–58 tok/s | Fits42–59 tok/s | Fits78–111 tok/s | | [solar-pro-preview-instruct](https://whatcanirunai.vercel.app/models/solar-pro-preview-instruct/)16.4 GB at Q4 | Fits42–59 tok/s | Fits43–61 tok/s | Fits80–113 tok/s | | [ERNIE-4.5-21B-A3B-PT](https://whatcanirunai.vercel.app/models/ernie-4.5-21b-a3b-pt/)16.3 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [ERNIE-4.5-21B-A3B-Thinking](https://whatcanirunai.vercel.app/models/ernie-4.5-21b-a3b-thinking/)16.2 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [gpt-oss-20b](https://whatcanirunai.vercel.app/models/gpt-oss-20b/)15.6 GB at Q4 · MoE | Fitsup to 365 tok/s | Fitsup to 374 tok/s | Fitsup to 698 tok/s | | [reka-flash-3](https://whatcanirunai.vercel.app/models/reka-flash-3/)15.6 GB at Q4 | Fits44–63 tok/s | Fits45–64 tok/s | Fits85–120 tok/s | | [maple-preview](https://whatcanirunai.vercel.app/models/maple-preview/)15.1 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1345 tok/s | Fitsup to 2511 tok/s | | [internlm2\_5-20b-chat](https://whatcanirunai.vercel.app/models/internlm2-5-20b-chat/)14.8 GB at Q4 | Fits47–66 tok/s | Fits48–68 tok/s | Fits89–127 tok/s | | [Groot-19B-A3B-261023](https://whatcanirunai.vercel.app/models/groot-19b-a3b-261023/)14.5 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [Ling-Coder-lite](https://whatcanirunai.vercel.app/models/ling-coder-lite/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ring-lite-distill-preview](https://whatcanirunai.vercel.app/models/ring-lite-distill-preview/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ling-lite](https://whatcanirunai.vercel.app/models/ling-lite/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ling-lite-1.5-2507](https://whatcanirunai.vercel.app/models/ling-lite-1.5-2507/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ling-lite-1.5](https://whatcanirunai.vercel.app/models/ling-lite-1.5/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ring-lite](https://whatcanirunai.vercel.app/models/ring-lite/)12.7 GB at Q4 · MoE | Fitsup to 477 tok/s | Fitsup to 489 tok/s | Fitsup to 913 tok/s | | [Ring-mini-linear-2.0](https://whatcanirunai.vercel.app/models/ring-mini-linear-2.0/)12.5 GB at Q4 · MoE | Fitsup to 820 tok/s | Fitsup to 841 tok/s | Fitsup to 1571 tok/s | | [Kimi-VL-A3B-Instruct](https://whatcanirunai.vercel.app/models/kimi-vl-a3b-instruct/)12.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Kimi-VL-A3B-Thinking](https://whatcanirunai.vercel.app/models/kimi-vl-a3b-thinking/)12.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Kimi-VL-A3B-Thinking-2506](https://whatcanirunai.vercel.app/models/kimi-vl-a3b-thinking-2506/)12.4 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [LLaDA2.1-mini](https://whatcanirunai.vercel.app/models/llada2.1-mini/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [LLaDA2.0-mini](https://whatcanirunai.vercel.app/models/llada2.0-mini/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [Ling-mini-2.0](https://whatcanirunai.vercel.app/models/ling-mini-2.0/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [Ring-mini-2.0](https://whatcanirunai.vercel.app/models/ring-mini-2.0/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [LLaDA2.2-mini](https://whatcanirunai.vercel.app/models/llada2.2-mini/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [LLaDA2.0-mini-preview](https://whatcanirunai.vercel.app/models/llada2.0-mini-preview/)12.3 GB at Q4 · MoE | Fitsup to 938 tok/s | Fitsup to 962 tok/s | Fitsup to 1795 tok/s | | [Moonlight-16B-A3B-Instruct](https://whatcanirunai.vercel.app/models/moonlight-16b-a3b-instruct/)12.1 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [Moonlight-16B-A3B](https://whatcanirunai.vercel.app/models/moonlight-16b-a3b/)12.1 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 837 tok/s | | [Instella-MoE-16B-A3B-Think](https://whatcanirunai.vercel.app/models/instella-moe-16b-a3b-think/)12.1 GB at Q4 · MoE | Fitsup to 437 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [DeepSeek-Coder-V2-Lite-Instruct](https://whatcanirunai.vercel.app/models/deepseek-coder-v2-lite-instruct/)12.0 GB at Q4 · MoE | Fitsup to 547 tok/s | Fitsup to 561 tok/s | Fitsup to 1047 tok/s | | [kanana-1.5-15.7b-a3b-instruct](https://whatcanirunai.vercel.app/models/kanana-1.5-15.7b-a3b-instruct/)11.9 GB at Q4 · MoE | Fitsup to 438 tok/s | Fitsup to 449 tok/s | Fitsup to 838 tok/s | | [Phi-4-reasoning-vision-15B](https://whatcanirunai.vercel.app/models/phi-4-reasoning-vision-15b/)11.5 GB at Q4 | Fits61–87 tok/s | Fits63–89 tok/s | Fits117–166 tok/s | | [Apriel-Nemotron-15b-Thinker](https://whatcanirunai.vercel.app/models/apriel-nemotron-15b-thinker/)11.4 GB at Q4 | Fits62–88 tok/s | Fits63–90 tok/s | Fits118–168 tok/s | | [Apriel-1.5-15b-Thinker](https://whatcanirunai.vercel.app/models/apriel-1.5-15b-thinker/)11.4 GB at Q4 | Fits62–88 tok/s | Fits64–91 tok/s | Fits119–169 tok/s | | [Apriel-1.6-15b-Thinker](https://whatcanirunai.vercel.app/models/apriel-1.6-15b-thinker/)11.4 GB at Q4 | Fits62–88 tok/s | Fits64–91 tok/s | Fits119–169 tok/s | | [Carbon-X (DNA)](https://whatcanirunai.vercel.app/models/carbon-x/)11.3 GB at Q4 | Fits62–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen2.5-Coder-14B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-14b-instruct/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [DeepSeek-R1-Distill-Qwen-14B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-qwen-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen2.5-14B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-14b-instruct/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen2.5-14B](https://whatcanirunai.vercel.app/models/qwen2.5-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen2.5-14B-Instruct-1M](https://whatcanirunai.vercel.app/models/qwen2.5-14b-instruct-1m/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen2.5-Coder-14B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [Qwen3-14B](https://whatcanirunai.vercel.app/models/qwen3-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [JanusCoder-14B](https://whatcanirunai.vercel.app/models/januscoder-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [HyperCLOVAX-SEED-Think-14B](https://whatcanirunai.vercel.app/models/hyperclovax-seed-think-14b/)11.3 GB at Q4 | Fits63–89 tok/s | Fits64–91 tok/s | Fits120–170 tok/s | | [phi-4](https://whatcanirunai.vercel.app/models/phi-4/)11.2 GB at Q4 | Fits63–90 tok/s | Fits65–92 tok/s | Fits121–171 tok/s | | [Phi-4-reasoning](https://whatcanirunai.vercel.app/models/phi-4-reasoning/)11.2 GB at Q4 | Fits63–90 tok/s | Fits65–92 tok/s | Fits121–171 tok/s | | [Phi-4-reasoning-plus](https://whatcanirunai.vercel.app/models/phi-4-reasoning-plus/)11.2 GB at Q4 | Fits63–90 tok/s | Fits65–92 tok/s | Fits121–171 tok/s | | [OLMo-2-1124-13B-Instruct](https://whatcanirunai.vercel.app/models/olmo-2-1124-13b-instruct/)10.6 GB at Q4 | Fits68–96 tok/s | Fits69–98 tok/s | Fits129–183 tok/s | | [Mistral-Nemo-Instruct-2407](https://whatcanirunai.vercel.app/models/mistral-nemo-instruct-2407/)9.5 GB at Q4 | Fits76–107 tok/s | Fits78–110 tok/s | Fits145–205 tok/s | | [gemma-3-12b-it](https://whatcanirunai.vercel.app/models/gemma-3-12b-it/)9.5 GB at Q4 | Fits76–108 tok/s | Fits78–110 tok/s | Fits146–206 tok/s | | [gemma-3-12b-pt](https://whatcanirunai.vercel.app/models/gemma-3-12b-pt/)9.5 GB at Q4 | Fits76–108 tok/s | Fits78–110 tok/s | Fits146–206 tok/s | | [Mellum2.1-12B-A2.5B-Thinking](https://whatcanirunai.vercel.app/models/mellum2.1-12b-a2.5b-thinking/)9.5 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 538 tok/s | Fitsup to 1005 tok/s | | [Mellum2-12B-A2.5B-Thinking](https://whatcanirunai.vercel.app/models/mellum2-12b-a2.5b-thinking/)9.5 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 538 tok/s | Fitsup to 1005 tok/s | | [Mellum2-12B-A2.5B-Instruct](https://whatcanirunai.vercel.app/models/mellum2-12b-a2.5b-instruct/)9.5 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 538 tok/s | Fitsup to 1005 tok/s | | [Mellum2-12B-A2.5B-Thinking-SFT](https://whatcanirunai.vercel.app/models/mellum2-12b-a2.5b-thinking-sft/)9.5 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 538 tok/s | Fitsup to 1005 tok/s | | [Mellum2.1-12B-A2.5B-Thinking-MTP](https://whatcanirunai.vercel.app/models/mellum2.1-12b-a2.5b-thinking-mtp/)9.5 GB at Q4 · MoE | Fitsup to 525 tok/s | Fitsup to 538 tok/s | Fitsup to 1005 tok/s | | [Gemma 4 12B](https://whatcanirunai.vercel.app/models/gemma-4-12b-it/)9.3 GB at Q4 | Fits77–110 tok/s | Fits79–113 tok/s | Fits148–210 tok/s | | [HyperCLOVAX-SEED-Omni-8B](https://whatcanirunai.vercel.app/models/hyperclovax-seed-omni-8b/)8.5 GB at Q4 | Fits86–122 tok/s | Fits88–125 tok/s | Fits165–234 tok/s | | [Llama-3.2-11B-Vision-Instruct](https://whatcanirunai.vercel.app/models/llama-3.2-11b-vision-instruct/)8.4 GB at Q4 | Fits87–123 tok/s | Fits89–126 tok/s | Fits166–235 tok/s | | [Llama-3.2-11B-Vision](https://whatcanirunai.vercel.app/models/llama-3.2-11b-vision/)8.4 GB at Q4 | Fits87–123 tok/s | Fits89–126 tok/s | Fits167–236 tok/s | | [Falcon3-10B-Instruct](https://whatcanirunai.vercel.app/models/falcon3-10b-instruct/)8.2 GB at Q4 | Fits90–127 tok/s | Fits92–131 tok/s | Fits172–244 tok/s | | [GLM-4.1V-9B-Thinking](https://whatcanirunai.vercel.app/models/glm-4.1v-9b-thinking/)8.2 GB at Q4 | Fits90–128 tok/s | Fits92–131 tok/s | Fits172–244 tok/s | | [GLM-4.6V-Flash](https://whatcanirunai.vercel.app/models/glm-4.6v-flash/)8.2 GB at Q4 | Fits90–128 tok/s | Fits92–131 tok/s | Fits172–244 tok/s | | [GLM-4.1V-9B-Base](https://whatcanirunai.vercel.app/models/glm-4.1v-9b-base/)8.2 GB at Q4 | Fits90–128 tok/s | Fits92–131 tok/s | Fits172–244 tok/s | | [Glyph](https://whatcanirunai.vercel.app/models/glyph/)8.2 GB at Q4 | Fits90–128 tok/s | Fits92–131 tok/s | Fits172–244 tok/s | | [INTELLECT-1-Instruct](https://whatcanirunai.vercel.app/models/intellect-1-instruct/)8.1 GB at Q4 | Fits91–129 tok/s | Fits93–132 tok/s | Fits174–246 tok/s | | [INTELLECT-1](https://whatcanirunai.vercel.app/models/intellect-1/)8.1 GB at Q4 | Fits91–129 tok/s | Fits93–132 tok/s | Fits174–246 tok/s | | [ZAYA1-VL-8B](https://whatcanirunai.vercel.app/models/zaya1-vl-8b/)7.8 GB at Q4 | Fits95–135 tok/s | Fits98–138 tok/s | Fits182–258 tok/s | | [SenseNova-Loop-9B-Preview](https://whatcanirunai.vercel.app/models/sensenova-loop-9b-preview/)7.7 GB at Q4 | Fits96–136 tok/s | Fits98–139 tok/s | Fits184–260 tok/s | | [Ornith-1.5-9B](https://whatcanirunai.vercel.app/models/ornith-1.5-9b/)7.7 GB at Q4 | Fits96–136 tok/s | Fits98–139 tok/s | Fits184–260 tok/s | | [Qwen3.5-9B](https://whatcanirunai.vercel.app/models/qwen3.5-9b/)7.7 GB at Q4 | Fits96–136 tok/s | Fits98–139 tok/s | Fits184–260 tok/s | | [MiniCPM-SALA](https://whatcanirunai.vercel.app/models/minicpm-sala/)7.6 GB at Q4 | Fits98–138 tok/s | Fits100–142 tok/s | Fits187–265 tok/s | | [NorthCUA-Nano](https://whatcanirunai.vercel.app/models/northcua-nano/)7.6 GB at Q4 | Fits98–139 tok/s | Fits101–143 tok/s | Fits188–267 tok/s | | [GLM-4-9B-0414](https://whatcanirunai.vercel.app/models/glm-4-9b-0414/)7.6 GB at Q4 | Fits99–140 tok/s | Fits101–143 tok/s | Fits189–267 tok/s | | [GLM-Z1-9B-0414](https://whatcanirunai.vercel.app/models/glm-z1-9b-0414/)7.6 GB at Q4 | Fits99–140 tok/s | Fits101–143 tok/s | Fits189–267 tok/s | | [codegeex4-all-9b](https://whatcanirunai.vercel.app/models/codegeex4-all-9b/)7.6 GB at Q4 | Fits99–140 tok/s | Fits101–143 tok/s | Fits189–267 tok/s | | [glm-4-9b-chat-hf](https://whatcanirunai.vercel.app/models/glm-4-9b-chat-hf/)7.6 GB at Q4 | Fits99–140 tok/s | Fits101–143 tok/s | Fits189–267 tok/s | | [moondream3-preview](https://whatcanirunai.vercel.app/models/moondream3-preview/)7.5 GB at Q4 | Fits100–142 tok/s | Fits103–145 tok/s | Fits191–271 tok/s | | [gemma-2-9b-it](https://whatcanirunai.vercel.app/models/gemma-2-9b-it/)7.4 GB at Q4 | Fits100–142 tok/s | Fits103–146 tok/s | Fits192–272 tok/s | | [gemma-2-9b](https://whatcanirunai.vercel.app/models/gemma-2-9b/)7.4 GB at Q4 | Fits100–142 tok/s | Fits103–146 tok/s | Fits192–272 tok/s | | [Gemma-SEA-LION-v3-9B-IT](https://whatcanirunai.vercel.app/models/gemma-sea-lion-v3-9b-it/)7.4 GB at Q4 | Fits100–142 tok/s | Fits103–146 tok/s | Fits192–272 tok/s | | [EuroLLM-9B-Instruct-2512](https://whatcanirunai.vercel.app/models/eurollm-9b-instruct-2512/)7.4 GB at Q4 | Fits101–143 tok/s | Fits104–147 tok/s | Fits194–275 tok/s | | [NVIDIA-Nemotron-Nano-9B-v2](https://whatcanirunai.vercel.app/models/nvidia-nemotron-nano-9b-v2/)7.2 GB at Q4 | Fits104–148 tok/s | Fits107–151 tok/s | Fits200–283 tok/s | | [ZAYA1-base](https://whatcanirunai.vercel.app/models/zaya1-base/)7.2 GB at Q4 | Fits105–148 tok/s | Fits107–152 tok/s | Fits201–284 tok/s | | [Yi-Coder-9B-Chat](https://whatcanirunai.vercel.app/models/yi-coder-9b-chat/)7.2 GB at Q4 | Fits105–149 tok/s | Fits108–152 tok/s | Fits201–285 tok/s | | [Yi-Coder-9B](https://whatcanirunai.vercel.app/models/yi-coder-9b/)7.2 GB at Q4 | Fits105–149 tok/s | Fits108–152 tok/s | Fits201–285 tok/s | | [internlm3-8b-instruct](https://whatcanirunai.vercel.app/models/internlm3-8b-instruct/)7.1 GB at Q4 | Fits105–149 tok/s | Fits108–153 tok/s | Fits201–285 tok/s | | [granite-4.1-8b](https://whatcanirunai.vercel.app/models/granite-4.1-8b/)7.1 GB at Q4 | Fits105–149 tok/s | Fits108–153 tok/s | Fits202–286 tok/s | | [granite-4.2-8b](https://whatcanirunai.vercel.app/models/granite-4.2-8b/)7.1 GB at Q4 | Fits105–149 tok/s | Fits108–153 tok/s | Fits202–286 tok/s | | [granite-switch-4.1-8b](https://whatcanirunai.vercel.app/models/granite-switch-4.1-8b/)7.1 GB at Q4 | Fits106–150 tok/s | Fits108–153 tok/s | Fits202–286 tok/s | | [Qwen3-VL-8B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-8b-instruct/)7.1 GB at Q4 | Fits106–150 tok/s | Fits108–154 tok/s | Fits202–287 tok/s | | [Qwen3-VL-8B-Thinking](https://whatcanirunai.vercel.app/models/qwen3-vl-8b-thinking/)7.1 GB at Q4 | Fits106–150 tok/s | Fits108–154 tok/s | Fits202–287 tok/s | | [UI-Venus-1.5-8B](https://whatcanirunai.vercel.app/models/ui-venus-1.5-8b/)7.1 GB at Q4 | Fits106–150 tok/s | Fits108–154 tok/s | Fits202–287 tok/s | | [Llama-3.1-Nemotron-Nano-VL-8B-V1](https://whatcanirunai.vercel.app/models/llama-3.1-nemotron-nano-vl-8b-v1/)7.1 GB at Q4 | Fits106–150 tok/s | Fits109–154 tok/s | Fits203–288 tok/s | | [MiniCPM-V-4\_5](https://whatcanirunai.vercel.app/models/minicpm-v-4-5/)7.1 GB at Q4 | Fits107–151 tok/s | Fits109–155 tok/s | Fits204–289 tok/s | | [aya-vision-8b](https://whatcanirunai.vercel.app/models/aya-vision-8b/)7.0 GB at Q4 | Fits107–152 tok/s | Fits110–156 tok/s | Fits205–291 tok/s | | [Intern-S1-mini](https://whatcanirunai.vercel.app/models/intern-s1-mini/)7.0 GB at Q4 | Fits108–154 tok/s | Fits111–158 tok/s | Fits208–294 tok/s | | [Nemotron-Labs-Diffusion-8B-Base](https://whatcanirunai.vercel.app/models/nemotron-labs-diffusion-8b-base/)6.9 GB at Q4 | Fits109–155 tok/s | Fits112–159 tok/s | Fits209–296 tok/s | | [LFM2.5-8B-A1B](https://whatcanirunai.vercel.app/models/lfm2.5-8b-a1b/)6.9 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2512 tok/s | | [Mistral-NeMo-Minitron-8B-Base](https://whatcanirunai.vercel.app/models/mistral-nemo-minitron-8b-base/)6.9 GB at Q4 | Fits110–156 tok/s | Fits113–160 tok/s | Fits211–299 tok/s | | [LFM2-8B-A1B](https://whatcanirunai.vercel.app/models/lfm2-8b-a1b/)6.8 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2513 tok/s | | [rnj-1-instruct](https://whatcanirunai.vercel.app/models/rnj-1-instruct/)6.8 GB at Q4 | Fits111–158 tok/s | Fits114–162 tok/s | Fits213–302 tok/s | | [MiMo-VL-7B-RL-2508](https://whatcanirunai.vercel.app/models/mimo-vl-7b-rl-2508/)6.8 GB at Q4 | Fits112–158 tok/s | Fits114–162 tok/s | Fits214–302 tok/s | | [MiMo-VL-7B-RL](https://whatcanirunai.vercel.app/models/mimo-vl-7b-rl/)6.8 GB at Q4 | Fits112–158 tok/s | Fits114–162 tok/s | Fits214–302 tok/s | | [MiMo-Embodied-7B](https://whatcanirunai.vercel.app/models/mimo-embodied-7b/)6.8 GB at Q4 | Fits112–158 tok/s | Fits114–162 tok/s | Fits214–302 tok/s | | [MiMo-VL-7B-SFT](https://whatcanirunai.vercel.app/models/mimo-vl-7b-sft/)6.8 GB at Q4 | Fits112–158 tok/s | Fits114–162 tok/s | Fits214–302 tok/s | | [Qwen2.5-VL-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-vl-7b-instruct/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [UI-TARS-1.5-7B](https://whatcanirunai.vercel.app/models/ui-tars-1.5-7b/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [Fara-7B](https://whatcanirunai.vercel.app/models/fara-7b/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [UI-Venus-Ground-7B](https://whatcanirunai.vercel.app/models/ui-venus-ground-7b/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [Qwen2-VL-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-vl-7b-instruct/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [UI-TARS-7B-DPO](https://whatcanirunai.vercel.app/models/ui-tars-7b-dpo/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [UI-TARS-7B-SFT](https://whatcanirunai.vercel.app/models/ui-tars-7b-sft/)6.8 GB at Q4 | Fits112–158 tok/s | Fits115–162 tok/s | Fits214–303 tok/s | | [Seed-Coder-8B-Instruct](https://whatcanirunai.vercel.app/models/seed-coder-8b-instruct/)6.8 GB at Q4 | Fits112–159 tok/s | Fits115–163 tok/s | Fits215–305 tok/s | | [Seed-Coder-8B-Reasoning](https://whatcanirunai.vercel.app/models/seed-coder-8b-reasoning/)6.8 GB at Q4 | Fits112–159 tok/s | Fits115–163 tok/s | Fits215–305 tok/s | | [Stable-DiffCoder-8B-Instruct](https://whatcanirunai.vercel.app/models/stable-diffcoder-8b-instruct/)6.8 GB at Q4 | Fits112–159 tok/s | Fits115–163 tok/s | Fits215–305 tok/s | | [Qwen3-8B](https://whatcanirunai.vercel.app/models/qwen3-8b/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [DeepSeek-R1-0528-Qwen3-8B](https://whatcanirunai.vercel.app/models/deepseek-r1-0528-qwen3-8b/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [Qwen3-8B-Base](https://whatcanirunai.vercel.app/models/qwen3-8b-base/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [WeDLM-8B-Instruct](https://whatcanirunai.vercel.app/models/wedlm-8b-instruct/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [PaCoRe-8B](https://whatcanirunai.vercel.app/models/pacore-8b/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [MiniCPM4-8B](https://whatcanirunai.vercel.app/models/minicpm4-8b/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [MiniCPM4.1-8B](https://whatcanirunai.vercel.app/models/minicpm4.1-8b/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [AgentCPM-Report](https://whatcanirunai.vercel.app/models/agentcpm-report/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [MiniCPM4-MCP](https://whatcanirunai.vercel.app/models/minicpm4-mcp/)6.7 GB at Q4 | Fits113–160 tok/s | Fits116–164 tok/s | Fits217–307 tok/s | | [granite-3.3-8b-instruct](https://whatcanirunai.vercel.app/models/granite-3.3-8b-instruct/)6.7 GB at Q4 | Fits113–161 tok/s | Fits116–165 tok/s | Fits217–307 tok/s | | [granite-3.1-8b-instruct](https://whatcanirunai.vercel.app/models/granite-3.1-8b-instruct/)6.7 GB at Q4 | Fits113–161 tok/s | Fits116–165 tok/s | Fits217–307 tok/s | | [granite-3.2-8b-instruct](https://whatcanirunai.vercel.app/models/granite-3.2-8b-instruct/)6.7 GB at Q4 | Fits113–161 tok/s | Fits116–165 tok/s | Fits217–307 tok/s | | [granite-3.0-8b-instruct](https://whatcanirunai.vercel.app/models/granite-3.0-8b-instruct/)6.7 GB at Q4 | Fits113–161 tok/s | Fits116–165 tok/s | Fits217–307 tok/s | | [Apertus-8B-Instruct-2509](https://whatcanirunai.vercel.app/models/apertus-8b-instruct-2509/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–167 tok/s | Fits220–312 tok/s | | [Apertus-8B-2509](https://whatcanirunai.vercel.app/models/apertus-8b-2509/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–167 tok/s | Fits220–312 tok/s | | [Llama-3.1-Tulu-3-8B](https://whatcanirunai.vercel.app/models/llama-3.1-tulu-3-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Llama-3.1-Tulu-3-8B-DPO](https://whatcanirunai.vercel.app/models/llama-3.1-tulu-3-8b-dpo/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Llama-3.1-Tulu-3.1-8B](https://whatcanirunai.vercel.app/models/llama-3.1-tulu-3.1-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [kanana-1.5-8b-instruct-2505](https://whatcanirunai.vercel.app/models/kanana-1.5-8b-instruct-2505/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Llama-3.1-8B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.1-8b-instruct/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Llama-3.1-8B](https://whatcanirunai.vercel.app/models/llama-3.1-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [DeepSeek-R1-Distill-Llama-8B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-llama-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Meta-Llama-3.1-8B-Instruct](https://whatcanirunai.vercel.app/models/meta-llama-3.1-8b-instruct/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [L3-8B-Stheno-v3.2](https://whatcanirunai.vercel.app/models/l3-8b-stheno-v3.2/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Llama-3.1-Nemotron-Nano-8B-v1](https://whatcanirunai.vercel.app/models/llama-3.1-nemotron-nano-8b-v1/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Meta-Llama-3.1-8B](https://whatcanirunai.vercel.app/models/meta-llama-3.1-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [LongWriter-llama3.1-8b](https://whatcanirunai.vercel.app/models/longwriter-llama3.1-8b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Hy-MT2-7B](https://whatcanirunai.vercel.app/models/hy-mt2-7b/)6.6 GB at Q4 | Fits115–163 tok/s | Fits118–168 tok/s | Fits221–313 tok/s | | [Nemotron 3.5 Bolt E4B](https://whatcanirunai.vercel.app/models/nemotron-3.5-bolt-e4b/)6.6 GB at Q4 | Fits116–164 tok/s | Fits119–168 tok/s | Fits222–314 tok/s | | [Gemma 4 E4B](https://whatcanirunai.vercel.app/models/gemma-4-e4b-it/)6.6 GB at Q4 | Fits116–164 tok/s | Fits119–168 tok/s | Fits222–314 tok/s | | [Ling-3.0-tiny](https://whatcanirunai.vercel.app/models/ling-3.0-tiny/)6.5 GB at Q4 · MoE | Fitsup to 1009 tok/s | Fitsup to 1035 tok/s | Fitsup to 1933 tok/s | | [MiMo-7B-RL](https://whatcanirunai.vercel.app/models/mimo-7b-rl/)6.5 GB at Q4 | Fits118–168 tok/s | Fits121–172 tok/s | Fits226–321 tok/s | | [MiMo-7B-Base](https://whatcanirunai.vercel.app/models/mimo-7b-base/)6.5 GB at Q4 | Fits118–168 tok/s | Fits121–172 tok/s | Fits226–321 tok/s | | [MiMo-7B-RL-0530](https://whatcanirunai.vercel.app/models/mimo-7b-rl-0530/)6.5 GB at Q4 | Fits118–168 tok/s | Fits121–172 tok/s | Fits226–321 tok/s | | [MiMo-7B-SFT](https://whatcanirunai.vercel.app/models/mimo-7b-sft/)6.5 GB at Q4 | Fits118–168 tok/s | Fits121–172 tok/s | Fits226–321 tok/s | | [EXAONE-3.5-7.8B-Instruct](https://whatcanirunai.vercel.app/models/exaone-3.5-7.8b-instruct/)6.5 GB at Q4 | Fits118–168 tok/s | Fits122–172 tok/s | Fits227–321 tok/s | | [EXAONE-Deep-7.8B](https://whatcanirunai.vercel.app/models/exaone-deep-7.8b/)6.5 GB at Q4 | Fits118–168 tok/s | Fits122–172 tok/s | Fits227–321 tok/s | | [salamandra-7b-instruct](https://whatcanirunai.vercel.app/models/salamandra-7b-instruct/)6.4 GB at Q4 | Fits119–169 tok/s | Fits122–173 tok/s | Fits228–323 tok/s | | [salamandra-7b](https://whatcanirunai.vercel.app/models/salamandra-7b/)6.4 GB at Q4 | Fits119–169 tok/s | Fits122–173 tok/s | Fits228–323 tok/s | | [salamandra-7b-instruct-2606](https://whatcanirunai.vercel.app/models/salamandra-7b-instruct-2606/)6.4 GB at Q4 | Fits119–169 tok/s | Fits122–173 tok/s | Fits228–323 tok/s | | [FastVLM-7B](https://whatcanirunai.vercel.app/models/fastvlm-7b/)6.4 GB at Q4 | Fits119–169 tok/s | Fits122–173 tok/s | Fits228–324 tok/s | | [Molmo2-O-7B](https://whatcanirunai.vercel.app/models/molmo2-o-7b/)6.4 GB at Q4 | Fits119–169 tok/s | Fits122–173 tok/s | Fits229–324 tok/s | | [internlm2\_5-7b-chat](https://whatcanirunai.vercel.app/models/internlm2-5-7b-chat/)6.4 GB at Q4 | Fits120–170 tok/s | Fits123–174 tok/s | Fits229–325 tok/s | | [internlm2\_5-7b-chat-1m](https://whatcanirunai.vercel.app/models/internlm2-5-7b-chat-1m/)6.4 GB at Q4 | Fits120–170 tok/s | Fits123–174 tok/s | Fits229–325 tok/s | | [Phi-mini-MoE-instruct](https://whatcanirunai.vercel.app/models/phi-mini-moe-instruct/)6.3 GB at Q4 · MoE | Fitsup to 547 tok/s | Fitsup to 561 tok/s | Fitsup to 1047 tok/s | | [Qwen2.5-Coder-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-7b-instruct/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [DeepSeek-R1-Distill-Qwen-7B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-qwen-7b/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-7b-instruct/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-7B](https://whatcanirunai.vercel.app/models/qwen2.5-7b/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-7b-instruct/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-Coder-7B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-7b/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2-7B](https://whatcanirunai.vercel.app/models/qwen2-7b/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-7B-Instruct-1M](https://whatcanirunai.vercel.app/models/qwen2.5-7b-instruct-1m/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-Math-7B](https://whatcanirunai.vercel.app/models/qwen2.5-math-7b/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Qwen2.5-Math-7B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-math-7b-instruct/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Marco-o1](https://whatcanirunai.vercel.app/models/marco-o1/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [SeaLLMs-v3-7B-Chat](https://whatcanirunai.vercel.app/models/seallms-v3-7b-chat/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Arcee-Spark](https://whatcanirunai.vercel.app/models/arcee-spark/)6.3 GB at Q4 | Fits122–172 tok/s | Fits125–177 tok/s | Fits233–330 tok/s | | [Falcon-H1-7B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-7b-instruct/)6.3 GB at Q4 | Fits122–173 tok/s | Fits125–177 tok/s | Fits234–331 tok/s | | [Falcon-H1R-7B](https://whatcanirunai.vercel.app/models/falcon-h1r-7b/)6.3 GB at Q4 | Fits122–173 tok/s | Fits125–177 tok/s | Fits234–331 tok/s | | [starvector-8b-im2svg](https://whatcanirunai.vercel.app/models/starvector-8b-im2svg/)6.2 GB at Q4 | Fits123–175 tok/s | Fits127–179 tok/s | Fits236–335 tok/s | | [Falcon3-7B-Instruct](https://whatcanirunai.vercel.app/models/falcon3-7b-instruct/)6.2 GB at Q4 | Fits124–176 tok/s | Fits127–181 tok/s | Fits238–337 tok/s | | [Olmo-Hybrid-7B](https://whatcanirunai.vercel.app/models/olmo-hybrid-7b/)6.2 GB at Q4 | Fits125–177 tok/s | Fits128–181 tok/s | Fits239–338 tok/s | | [LLaDA-MoE-7B-A1B-Base](https://whatcanirunai.vercel.app/models/llada-moe-7b-a1b-base/)6.1 GB at Q4 · MoE | Fitsup to 1313 tok/s | Fitsup to 1346 tok/s | Fitsup to 2513 tok/s | | [Zamba2-7B-Instruct](https://whatcanirunai.vercel.app/models/zamba2-7b-instruct/)6.1 GB at Q4 | Fits126–178 tok/s | Fits129–183 tok/s | Fits241–342 tok/s | | [OLMo-2-1124-7B-Instruct](https://whatcanirunai.vercel.app/models/olmo-2-1124-7b-instruct/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Think](https://whatcanirunai.vercel.app/models/olmo-3-7b-think/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Instruct](https://whatcanirunai.vercel.app/models/olmo-3-7b-instruct/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-1025-7B](https://whatcanirunai.vercel.app/models/olmo-3-1025-7b/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Instruct-SFT](https://whatcanirunai.vercel.app/models/olmo-3-7b-instruct-sft/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Think-SFT](https://whatcanirunai.vercel.app/models/olmo-3-7b-think-sft/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Instruct-DPO](https://whatcanirunai.vercel.app/models/olmo-3-7b-instruct-dpo/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-RL-Zero-Math](https://whatcanirunai.vercel.app/models/olmo-3-7b-rl-zero-math/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3.1-7B-RL-Zero-Math](https://whatcanirunai.vercel.app/models/olmo-3.1-7b-rl-zero-math/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Olmo-3-7B-Think-DPO](https://whatcanirunai.vercel.app/models/olmo-3-7b-think-dpo/)6.1 GB at Q4 | Fits127–180 tok/s | Fits130–184 tok/s | Fits243–344 tok/s | | [Mistral-7B-Instruct-v0.3](https://whatcanirunai.vercel.app/models/mistral-7b-instruct-v0.3/)6.1 GB at Q4 | Fits128–181 tok/s | Fits131–186 tok/s | Fits245–347 tok/s | | [granite-4.0-h-tiny](https://whatcanirunai.vercel.app/models/granite-4.0-h-tiny/)5.8 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2513 tok/s | | [OLMoE-1B-7B-0924-Instruct](https://whatcanirunai.vercel.app/models/olmoe-1b-7b-0924-instruct/)5.8 GB at Q4 · MoE | Fitsup to 1009 tok/s | Fitsup to 1035 tok/s | Fitsup to 1933 tok/s | | [OLMoE-1B-7B-0924](https://whatcanirunai.vercel.app/models/olmoe-1b-7b-0924/)5.8 GB at Q4 · MoE | Fitsup to 1009 tok/s | Fitsup to 1035 tok/s | Fitsup to 1933 tok/s | | [OLMoE-1B-7B-0125-Instruct](https://whatcanirunai.vercel.app/models/olmoe-1b-7b-0125-instruct/)5.8 GB at Q4 · MoE | Fitsup to 1009 tok/s | Fitsup to 1035 tok/s | Fitsup to 1933 tok/s | | [OLMoE-1B-7B-0125](https://whatcanirunai.vercel.app/models/olmoe-1b-7b-0125/)5.8 GB at Q4 · MoE | Fitsup to 1009 tok/s | Fitsup to 1035 tok/s | Fitsup to 1933 tok/s | | [OLMo-7B-0724-hf](https://whatcanirunai.vercel.app/models/olmo-7b-0724-hf/)5.8 GB at Q4 | Fits134–191 tok/s | Fits138–195 tok/s | Fits257–365 tok/s | | [granite-4.0-tiny-preview](https://whatcanirunai.vercel.app/models/granite-4.0-tiny-preview/)5.7 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2512 tok/s | | [YuE-s1-7B-anneal-en-cot](https://whatcanirunai.vercel.app/models/yue-s1-7b-anneal-en-cot/)5.3 GB at Q4 | Fits149–211 tok/s | Fits153–216 tok/s | Fits285–404 tok/s | | [Trinity-Nano-Preview](https://whatcanirunai.vercel.app/models/trinity-nano-preview/)5.3 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2512 tok/s | | [Trinity-Nano-Base](https://whatcanirunai.vercel.app/models/trinity-nano-base/)5.3 GB at Q4 · MoE | Fitsup to 1312 tok/s | Fitsup to 1346 tok/s | Fitsup to 2512 tok/s | | [gemma-3n-E2B-it](https://whatcanirunai.vercel.app/models/gemma-3n-e2b-it/)4.8 GB at Q4 | Fits170–241 tok/s | Fits175–247 tok/s | Fits326–462 tok/s | | [Youtu-VL-4B-Instruct](https://whatcanirunai.vercel.app/models/youtu-vl-4b-instruct/)4.7 GB at Q4 | Fits173–246 tok/s | Fits178–252 tok/s | Fits332–470 tok/s | | [Gemma 4 E2B](https://whatcanirunai.vercel.app/models/gemma-4-e2b-it/)4.6 GB at Q4 | Fits181–256 tok/s | Fits185–263 tok/s | Fits346–490 tok/s | | [Molmo2-ER](https://whatcanirunai.vercel.app/models/molmo2-er/)4.4 GB at Q4 | Fits191–271 tok/s | Fits196–277 tok/s | Fits366–518 tok/s | | [Apriel-5B-Instruct](https://whatcanirunai.vercel.app/models/apriel-5b-instruct/)4.4 GB at Q4 | Fits192–272 tok/s | Fits197–279 tok/s | Fits367–520 tok/s | | [WeVisOne-VL-4B](https://whatcanirunai.vercel.app/models/wevisone-vl-4b/)4.4 GB at Q4 | Fits192–272 tok/s | Fits197–279 tok/s | Fits367–520 tok/s | | [Mage-VL](https://whatcanirunai.vercel.app/models/mage-vl/)4.3 GB at Q4 | Fits195–277 tok/s | Fits200–284 tok/s | Fits374–530 tok/s | | [Mage-VL-Cowork](https://whatcanirunai.vercel.app/models/mage-vl-cowork/)4.3 GB at Q4 | Fits195–277 tok/s | Fits200–284 tok/s | Fits374–530 tok/s | | [Qwen3.5-4B](https://whatcanirunai.vercel.app/models/qwen3.5-4b/)4.2 GB at Q4 | Fits199–282 tok/s | Fits204–289 tok/s | Fits381–539 tok/s | | [S5-nano](https://whatcanirunai.vercel.app/models/s5-nano/)4.2 GB at Q4 | Fits199–282 tok/s | Fits204–289 tok/s | Fits381–539 tok/s | | [AFM-4.5B](https://whatcanirunai.vercel.app/models/afm-4.5b/)4.2 GB at Q4 | Fits201–284 tok/s | Fits206–291 tok/s | Fits384–544 tok/s | | [Llama-3.1-Minitron-4B-Width-Base](https://whatcanirunai.vercel.app/models/llama-3.1-minitron-4b-width-base/)4.1 GB at Q4 | Fits205–291 tok/s | Fits211–298 tok/s | Fits393–557 tok/s | | [Qwen3-VL-4B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-4b-instruct/)4.1 GB at Q4 | Fits209–296 tok/s | Fits214–303 tok/s | Fits400–566 tok/s | | [CapRL-Qwen3VL-4B](https://whatcanirunai.vercel.app/models/caprl-qwen3vl-4b/)4.1 GB at Q4 | Fits209–296 tok/s | Fits214–303 tok/s | Fits400–566 tok/s | | [glm-edge-4b-chat](https://whatcanirunai.vercel.app/models/glm-edge-4b-chat/)4.0 GB at Q4 | Fits214–303 tok/s | Fits220–311 tok/s | Fits410–581 tok/s | | [gemma-3-4b-it](https://whatcanirunai.vercel.app/models/gemma-3-4b-it/)4.0 GB at Q4 | Fits215–305 tok/s | Fits221–313 tok/s | Fits412–584 tok/s | | [medgemma-4b-it](https://whatcanirunai.vercel.app/models/medgemma-4b-it/)4.0 GB at Q4 | Fits215–305 tok/s | Fits221–313 tok/s | Fits412–584 tok/s | | [medgemma-1.5-4b-it](https://whatcanirunai.vercel.app/models/medgemma-1.5-4b-it/)4.0 GB at Q4 | Fits215–305 tok/s | Fits221–313 tok/s | Fits412–584 tok/s | | [gemma-3-4b-pt](https://whatcanirunai.vercel.app/models/gemma-3-4b-pt/)4.0 GB at Q4 | Fits215–305 tok/s | Fits221–313 tok/s | Fits412–584 tok/s | | [Nanbeige4.2-3B](https://whatcanirunai.vercel.app/models/nanbeige4.2-3b/)3.9 GB at Q4 | Fits222–315 tok/s | Fits228–323 tok/s | Fits425–603 tok/s | | [Phi-3.5-vision-instruct](https://whatcanirunai.vercel.app/models/phi-3.5-vision-instruct/)3.9 GB at Q4 | Fits223–316 tok/s | Fits229–325 tok/s | Fits428–606 tok/s | | [Spark-X2.6-4B](https://whatcanirunai.vercel.app/models/spark-x2.6-4b/)3.9 GB at Q4 | Fits225–319 tok/s | Fits231–327 tok/s | Fits431–611 tok/s | | [Spark-X2.5-4B](https://whatcanirunai.vercel.app/models/spark-x2.5-4b/)3.9 GB at Q4 | Fits225–319 tok/s | Fits231–327 tok/s | Fits431–611 tok/s | | [POINTS-Reader](https://whatcanirunai.vercel.app/models/points-reader/)3.9 GB at Q4 | Fits225–319 tok/s | Fits231–328 tok/s | Fits432–611 tok/s | | [MiniCPM-V-4](https://whatcanirunai.vercel.app/models/minicpm-v-4/)3.8 GB at Q4 | Fits228–323 tok/s | Fits234–332 tok/s | Fits437–619 tok/s | | [Qwen3-4B](https://whatcanirunai.vercel.app/models/qwen3-4b/)3.8 GB at Q4 | Fits230–326 tok/s | Fits236–335 tok/s | Fits441–625 tok/s | | [Qwen3-4B-Instruct-2507](https://whatcanirunai.vercel.app/models/qwen3-4b-instruct-2507/)3.8 GB at Q4 | Fits230–326 tok/s | Fits236–335 tok/s | Fits441–625 tok/s | | [Qwen3-4B-Thinking-2507](https://whatcanirunai.vercel.app/models/qwen3-4b-thinking-2507/)3.8 GB at Q4 | Fits230–326 tok/s | Fits236–335 tok/s | Fits441–625 tok/s | | [Mellum-4b-base](https://whatcanirunai.vercel.app/models/mellum-4b-base/)3.8 GB at Q4 | Fits230–327 tok/s | Fits236–335 tok/s | Fits441–625 tok/s | | [granite-4.0-3b-vision](https://whatcanirunai.vercel.app/models/granite-4.0-3b-vision/)3.8 GB at Q4 | Fits232–328 tok/s | Fits238–337 tok/s | Fits444–629 tok/s | | [granite-vision-4.1-4b](https://whatcanirunai.vercel.app/models/granite-vision-4.1-4b/)3.8 GB at Q4 | Fits232–328 tok/s | Fits238–337 tok/s | Fits444–629 tok/s | | [Nemotron Labs Dory 1.0 4B](https://whatcanirunai.vercel.app/models/nemotron-labs-dory1.0-4b/)3.8 GB at Q4 | Fits233–330 tok/s | Fits239–339 tok/s | Fits446–632 tok/s | | [NVIDIA-Nemotron-3-Nano-4B-BF16](https://whatcanirunai.vercel.app/models/nvidia-nemotron-3-nano-4b-bf16/)3.8 GB at Q4 | Fits233–330 tok/s | Fits239–339 tok/s | Fits446–632 tok/s | | [Nanbeige4.1-3B](https://whatcanirunai.vercel.app/models/nanbeige4.1-3b/)3.7 GB at Q4 | Fits235–334 tok/s | Fits242–342 tok/s | Fits451–639 tok/s | | [Nanbeige4-3B-Thinking-2511](https://whatcanirunai.vercel.app/models/nanbeige4-3b-thinking-2511/)3.7 GB at Q4 | Fits235–334 tok/s | Fits242–342 tok/s | Fits451–639 tok/s | | [Phi-4-mini-flash-reasoning](https://whatcanirunai.vercel.app/models/phi-4-mini-flash-reasoning/)3.7 GB at Q4 | Fits240–341 tok/s | Fits247–349 tok/s | Fits460–652 tok/s | | [Phi-4-mini-instruct](https://whatcanirunai.vercel.app/models/phi-4-mini-instruct/)3.7 GB at Q4 | Fits241–342 tok/s | Fits248–351 tok/s | Fits462–655 tok/s | | [Phi-4-mini-reasoning](https://whatcanirunai.vercel.app/models/phi-4-mini-reasoning/)3.7 GB at Q4 | Fits241–342 tok/s | Fits248–351 tok/s | Fits462–655 tok/s | | [Nemotron-Labs-Diffusion-3B](https://whatcanirunai.vercel.app/models/nemotron-labs-diffusion-3b/)3.7 GB at Q4 | Fits242–342 tok/s | Fits248–351 tok/s | Fits463–656 tok/s | | [Phi-3.5-mini-instruct](https://whatcanirunai.vercel.app/models/phi-3.5-mini-instruct/)3.7 GB at Q4 | Fits242–343 tok/s | Fits249–352 tok/s | Fits464–658 tok/s | | [MediPhi-Instruct](https://whatcanirunai.vercel.app/models/mediphi-instruct/)3.7 GB at Q4 | Fits242–343 tok/s | Fits249–352 tok/s | Fits464–658 tok/s | | [MediPhi](https://whatcanirunai.vercel.app/models/mediphi/)3.7 GB at Q4 | Fits242–343 tok/s | Fits249–352 tok/s | Fits464–658 tok/s | | [MediPhi-Clinical](https://whatcanirunai.vercel.app/models/mediphi-clinical/)3.7 GB at Q4 | Fits242–343 tok/s | Fits249–352 tok/s | Fits464–658 tok/s | | [MediPhi-PubMed](https://whatcanirunai.vercel.app/models/mediphi-pubmed/)3.7 GB at Q4 | Fits242–343 tok/s | Fits249–352 tok/s | Fits464–658 tok/s | | [HY-Embodied-0.5](https://whatcanirunai.vercel.app/models/hy-embodied-0.5/)3.6 GB at Q4 | Fits245–347 tok/s | Fits251–356 tok/s | Fits469–664 tok/s | | [Phi-tiny-MoE-instruct](https://whatcanirunai.vercel.app/models/phi-tiny-moe-instruct/)3.6 GB at Q4 · MoE | Fitsup to 1193 tok/s | Fitsup to 1224 tok/s | Fitsup to 2284 tok/s | | [CapRL-3B](https://whatcanirunai.vercel.app/models/caprl-3b/)3.6 GB at Q4 | Fits247–350 tok/s | Fits253–358 tok/s | Fits472–669 tok/s | | [HyperCLOVAX-SEED-Vision-Instruct-3B](https://whatcanirunai.vercel.app/models/hyperclovax-seed-vision-instruct-3b/)3.6 GB at Q4 | Fits249–353 tok/s | Fits255–362 tok/s | Fits477–675 tok/s | | [kanana-1.5-v-3b-instruct](https://whatcanirunai.vercel.app/models/kanana-1.5-v-3b-instruct/)3.6 GB at Q4 | Fits253–358 tok/s | Fits259–367 tok/s | Fits484–685 tok/s | | [granite-4.2-3b](https://whatcanirunai.vercel.app/models/granite-4.2-3b/)3.6 GB at Q4 | Fits253–359 tok/s | Fits260–368 tok/s | Fits485–687 tok/s | | [kanana-2-3b-instruct](https://whatcanirunai.vercel.app/models/kanana-2-3b-instruct/)3.4 GB at Q4 | Fits264–374 tok/s | Fits271–384 tok/s | Fits505–716 tok/s | | [granite-4.1-3b](https://whatcanirunai.vercel.app/models/granite-4.1-3b/)3.4 GB at Q4 | Fits272–386 tok/s | Fits279–396 tok/s | Fits521–738 tok/s | | [granite-4.0-micro](https://whatcanirunai.vercel.app/models/granite-4.0-micro/)3.4 GB at Q4 | Fits272–386 tok/s | Fits279–396 tok/s | Fits521–738 tok/s | | [granite-3.1-3b-a800m-instruct](https://whatcanirunai.vercel.app/models/granite-3.1-3b-a800m-instruct/)3.3 GB at Q4 · MoE | Fitsup to 1640 tok/s | Fitsup to 1683 tok/s | Fitsup to 3141 tok/s | | [Falcon3-3B-Instruct](https://whatcanirunai.vercel.app/models/falcon3-3b-instruct/)3.3 GB at Q4 | Fits287–407 tok/s | Fits294–417 tok/s | Fits549–778 tok/s | | [Llama-3.2-3B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.2-3b-instruct/)3.2 GB at Q4 | Fits288–408 tok/s | Fits296–419 tok/s | Fits552–782 tok/s | | [Llama-3.2-3B](https://whatcanirunai.vercel.app/models/llama-3.2-3b/)3.2 GB at Q4 | Fits288–408 tok/s | Fits296–419 tok/s | Fits552–782 tok/s | | [Hermes-3-Llama-3.2-3B](https://whatcanirunai.vercel.app/models/hermes-3-llama-3.2-3b/)3.2 GB at Q4 | Fits288–408 tok/s | Fits296–419 tok/s | Fits552–782 tok/s | | [DeepHermes-3-Llama-3-3B-Preview](https://whatcanirunai.vercel.app/models/deephermes-3-llama-3-3b-preview/)3.2 GB at Q4 | Fits288–408 tok/s | Fits296–419 tok/s | Fits552–782 tok/s | | [AI21-Jamba-Reasoning-3B](https://whatcanirunai.vercel.app/models/ai21-jamba-reasoning-3b/)3.2 GB at Q4 | Fits290–410 tok/s | Fits297–421 tok/s | Fits555–786 tok/s | | [granite-4.0-h-micro](https://whatcanirunai.vercel.app/models/granite-4.0-h-micro/)3.2 GB at Q4 | Fits290–411 tok/s | Fits298–422 tok/s | Fits556–787 tok/s | | [Falcon-H1-3B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-3b-instruct/)3.2 GB at Q4 | Fits294–417 tok/s | Fits302–427 tok/s | Fits563–798 tok/s | | [d1-3B](https://whatcanirunai.vercel.app/models/d1-3b/)3.2 GB at Q4 | Fits297–420 tok/s | Fits304–431 tok/s | Fits568–804 tok/s | | [LFM2.5-VL-3B](https://whatcanirunai.vercel.app/models/lfm2.5-vl-3b/)3.2 GB at Q4 | Fits297–420 tok/s | Fits304–431 tok/s | Fits568–804 tok/s | | [Qwen2.5-3B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-3b-instruct/)3.2 GB at Q4 | Fits300–425 tok/s | Fits308–436 tok/s | Fits575–814 tok/s | | [Qwen2.5-Coder-3B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-3b-instruct/)3.2 GB at Q4 | Fits300–425 tok/s | Fits308–436 tok/s | Fits575–814 tok/s | | [Qwen2.5-3B](https://whatcanirunai.vercel.app/models/qwen2.5-3b/)3.2 GB at Q4 | Fits300–425 tok/s | Fits308–436 tok/s | Fits575–814 tok/s | | [Qwen2.5-Coder-3B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-3b/)3.2 GB at Q4 | Fits300–425 tok/s | Fits308–436 tok/s | Fits575–814 tok/s | | [SmolLM3-3B](https://whatcanirunai.vercel.app/models/smollm3-3b/)3.1 GB at Q4 | Fits301–427 tok/s | Fits309–438 tok/s | Fits577–817 tok/s | | [AI21-Jamba2-3B](https://whatcanirunai.vercel.app/models/ai21-jamba2-3b/)3.1 GB at Q4 | Fits306–433 tok/s | Fits314–444 tok/s | Fits585–829 tok/s | | [granite-swash-3b-a600m](https://whatcanirunai.vercel.app/models/granite-swash-3b-a600m/)3.1 GB at Q4 · MoE | Fitsup to 2194 tok/s | Fitsup to 2251 tok/s | Fitsup to 4201 tok/s | | [granite-vision-3.2-2b](https://whatcanirunai.vercel.app/models/granite-vision-3.2-2b/)3.1 GB at Q4 | Fits311–441 tok/s | Fits319–452 tok/s | Fits596–844 tok/s | | [granite-vision-3.1-2b-preview](https://whatcanirunai.vercel.app/models/granite-vision-3.1-2b-preview/)3.1 GB at Q4 | Fits311–441 tok/s | Fits319–452 tok/s | Fits596–844 tok/s | | [granite-vision-3.3-2b-chart2csv-preview](https://whatcanirunai.vercel.app/models/granite-vision-3.3-2b-chart2csv-preview/)3.1 GB at Q4 | Fits311–441 tok/s | Fits319–452 tok/s | Fits596–844 tok/s | | [LFM2.5-2.6B](https://whatcanirunai.vercel.app/models/lfm2.5-2.6b/)2.9 GB at Q4 | Fits343–487 tok/s | Fits352–499 tok/s | Fits658–932 tok/s | | [Zamba2-2.7B](https://whatcanirunai.vercel.app/models/zamba2-2.7b/)2.9 GB at Q4 | Fits348–493 tok/s | Fits357–506 tok/s | Fits666–944 tok/s | | [Zamba2-2.7B-instruct](https://whatcanirunai.vercel.app/models/zamba2-2.7b-instruct/)2.9 GB at Q4 | Fits348–493 tok/s | Fits357–506 tok/s | Fits666–944 tok/s | | [granite-3.0-2b-instruct](https://whatcanirunai.vercel.app/models/granite-3.0-2b-instruct/)2.8 GB at Q4 | Fits352–498 tok/s | Fits361–511 tok/s | Fits673–954 tok/s | | [gemma-2-2b-it](https://whatcanirunai.vercel.app/models/gemma-2-2b-it/)2.8 GB at Q4 | Fits354–502 tok/s | Fits363–515 tok/s | Fits678–961 tok/s | | [gemma-2-2b](https://whatcanirunai.vercel.app/models/gemma-2-2b/)2.8 GB at Q4 | Fits354–502 tok/s | Fits363–515 tok/s | Fits678–961 tok/s | | [EuroMoE-2.6B-A0.6B-Instruct-Preview](https://whatcanirunai.vercel.app/models/euromoe-2.6b-a0.6b-instruct-preview/)2.8 GB at Q4 · MoE | Fitsup to 2187 tok/s | Fitsup to 2243 tok/s | Fitsup to 4187 tok/s | | [Motif-2.6B](https://whatcanirunai.vercel.app/models/motif-2.6b/)2.8 GB at Q4 | Fits357–505 tok/s | Fits366–518 tok/s | Fits683–967 tok/s | | [LFM2-2.6B](https://whatcanirunai.vercel.app/models/lfm2-2.6b/)2.8 GB at Q4 | Fits361–511 tok/s | Fits370–524 tok/s | Fits690–978 tok/s | | [LFM2-2.6B-Exp](https://whatcanirunai.vercel.app/models/lfm2-2.6b-exp/)2.8 GB at Q4 | Fits361–511 tok/s | Fits370–524 tok/s | Fits690–978 tok/s | | [LFM2-2.6B-Transcript](https://whatcanirunai.vercel.app/models/lfm2-2.6b-transcript/)2.8 GB at Q4 | Fits361–511 tok/s | Fits370–524 tok/s | Fits690–978 tok/s | | [granite-3.3-2b-instruct](https://whatcanirunai.vercel.app/models/granite-3.3-2b-instruct/)2.8 GB at Q4 | Fits366–518 tok/s | Fits375–531 tok/s | Fits700–992 tok/s | | [granite-3.1-2b-instruct](https://whatcanirunai.vercel.app/models/granite-3.1-2b-instruct/)2.8 GB at Q4 | Fits366–518 tok/s | Fits375–531 tok/s | Fits700–992 tok/s | | [granite-3.2-2b-instruct](https://whatcanirunai.vercel.app/models/granite-3.2-2b-instruct/)2.8 GB at Q4 | Fits366–518 tok/s | Fits375–531 tok/s | Fits700–992 tok/s | | [MiniCPM5-2B](https://whatcanirunai.vercel.app/models/minicpm5-2b/)2.8 GB at Q4 | Fits368–521 tok/s | Fits378–535 tok/s | Fits705–998 tok/s | | [MiniCPM5-2B-SFT](https://whatcanirunai.vercel.app/models/minicpm5-2b-sft/)2.8 GB at Q4 | Fits368–521 tok/s | Fits378–535 tok/s | Fits705–998 tok/s | | [MiniCPM5-2B-Midtrain](https://whatcanirunai.vercel.app/models/minicpm5-2b-midtrain/)2.8 GB at Q4 | Fits368–521 tok/s | Fits378–535 tok/s | Fits705–998 tok/s | | [sarvam-1-v0.5](https://whatcanirunai.vercel.app/models/sarvam-1-v0.5/)2.7 GB at Q4 | Fits369–523 tok/s | Fits379–536 tok/s | Fits707–1001 tok/s | | [North-Micro-Vision-Instruct](https://whatcanirunai.vercel.app/models/north-micro-vision-instruct/)2.7 GB at Q4 | Fits373–528 tok/s | Fits382–542 tok/s | Fits714–1011 tok/s | | [UI-TARS-2B-SFT](https://whatcanirunai.vercel.app/models/ui-tars-2b-sft/)2.7 GB at Q4 | Fits379–537 tok/s | Fits389–551 tok/s | Fits726–1029 tok/s | | [WeVisOne-VL-2B](https://whatcanirunai.vercel.app/models/wevisone-vl-2b/)2.7 GB at Q4 | Fits379–537 tok/s | Fits389–551 tok/s | Fits726–1029 tok/s | | [UI-Venus-1.5-2B](https://whatcanirunai.vercel.app/models/ui-venus-1.5-2b/)2.7 GB at Q4 | Fits380–538 tok/s | Fits390–552 tok/s | Fits727–1030 tok/s | | [bitnet-b1.58-2B-4T-bf16](https://whatcanirunai.vercel.app/models/bitnet-b1.58-2b-4t-bf16/)2.7 GB at Q4 | Fits384–544 tok/s | Fits394–558 tok/s | Fits735–1041 tok/s | | [EXAONE-3.5-2.4B-Instruct](https://whatcanirunai.vercel.app/models/exaone-3.5-2.4b-instruct/)2.7 GB at Q4 | Fits385–546 tok/s | Fits395–560 tok/s | Fits737–1045 tok/s | | [EXAONE-Deep-2.4B](https://whatcanirunai.vercel.app/models/exaone-deep-2.4b/)2.7 GB at Q4 | Fits385–546 tok/s | Fits395–560 tok/s | Fits737–1045 tok/s | | [kanana-1.5-2.1b-instruct-2505](https://whatcanirunai.vercel.app/models/kanana-1.5-2.1b-instruct-2505/)2.6 GB at Q4 | Fits400–566 tok/s | Fits410–581 tok/s | Fits765–1084 tok/s | | [Qwen3.5-2B](https://whatcanirunai.vercel.app/models/qwen3.5-2b/)2.6 GB at Q4 | Fits407–577 tok/s | Fits418–592 tok/s | Fits780–1105 tok/s | | [salamandra-2b](https://whatcanirunai.vercel.app/models/salamandra-2b/)2.6 GB at Q4 | Fits411–582 tok/s | Fits422–597 tok/s | Fits787–1115 tok/s | | [salamandra-2b-instruct](https://whatcanirunai.vercel.app/models/salamandra-2b-instruct/)2.6 GB at Q4 | Fits411–582 tok/s | Fits422–597 tok/s | Fits787–1115 tok/s | | [SmolVLM2-2.2B-Instruct](https://whatcanirunai.vercel.app/models/smolvlm2-2.2b-instruct/)2.6 GB at Q4 | Fits412–584 tok/s | Fits423–599 tok/s | Fits789–1118 tok/s | | [SmolVLM-Base](https://whatcanirunai.vercel.app/models/smolvlm-base/)2.6 GB at Q4 | Fits412–584 tok/s | Fits423–599 tok/s | Fits790–1119 tok/s | | [Qwen2-VL-2B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-vl-2b-instruct/)2.5 GB at Q4 | Fits419–594 tok/s | Fits430–609 tok/s | Fits803–1137 tok/s | | [LFM3-Omni-2B](https://whatcanirunai.vercel.app/models/lfm3-omni-2b/)2.5 GB at Q4 | Fits427–606 tok/s | Fits438–621 tok/s | Fits818–1159 tok/s | | [granite-swash-2b](https://whatcanirunai.vercel.app/models/granite-swash-2b/)2.5 GB at Q4 | Fits432–612 tok/s | Fits443–628 tok/s | Fits827–1172 tok/s | | [Qwen3-VL-2B-Instruct](https://whatcanirunai.vercel.app/models/qwen3-vl-2b-instruct/)2.5 GB at Q4 | Fits435–617 tok/s | Fits447–633 tok/s | Fits834–1181 tok/s | | [kanana-nano-2.1b-instruct](https://whatcanirunai.vercel.app/models/kanana-nano-2.1b-instruct/)2.5 GB at Q4 | Fits444–629 tok/s | Fits455–645 tok/s | Fits850–1204 tok/s | | [Hy-MT2-1.8B](https://whatcanirunai.vercel.app/models/hy-mt2-1.8b/)2.4 GB at Q4 | Fits454–644 tok/s | Fits466–660 tok/s | Fits870–1233 tok/s | | [Qwen3-1.7B](https://whatcanirunai.vercel.app/models/qwen3-1.7b/)2.4 GB at Q4 | Fits456–646 tok/s | Fits468–662 tok/s | Fits873–1237 tok/s | | [Youtu-LLM-2B](https://whatcanirunai.vercel.app/models/youtu-llm-2b/)2.4 GB at Q4 | Fits472–669 tok/s | Fits484–686 tok/s | Fits904–1281 tok/s | | [LFM3-VL-2B](https://whatcanirunai.vercel.app/models/lfm3-vl-2b/)2.3 GB at Q4 | Fits483–684 tok/s | Fits495–702 tok/s | Fits925–1310 tok/s | | [d1-2B](https://whatcanirunai.vercel.app/models/d1-2b/)2.3 GB at Q4 | Fits483–684 tok/s | Fits495–702 tok/s | Fits925–1310 tok/s | | [FastVLM-1.5B](https://whatcanirunai.vercel.app/models/fastvlm-1.5b/)2.3 GB at Q4 | Fits485–687 tok/s | Fits498–705 tok/s | Fits929–1316 tok/s | | [internlm2\_5-1\_8b-chat](https://whatcanirunai.vercel.app/models/internlm2-5-1-8b-chat/)2.3 GB at Q4 | Fits490–695 tok/s | Fits503–712 tok/s | Fits939–1330 tok/s | | [DeepSeek-R1-Distill-Qwen-1.5B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-qwen-1.5b/)2.2 GB at Q4 | Fits521–738 tok/s | Fits535–757 tok/s | Fits998–1414 tok/s | | [LFM3-1.7B](https://whatcanirunai.vercel.app/models/lfm3-1.7b/)2.2 GB at Q4 | Fits536–759 tok/s | Fits550–779 tok/s | Fits1026–1453 tok/s | | [Qwen3-1.7B-Base](https://whatcanirunai.vercel.app/models/qwen3-1.7b-base/)2.2 GB at Q4 | Fits538–763 tok/s | Fits552–782 tok/s | Fits1031–1460 tok/s | | [SmolLM2-1.7B](https://whatcanirunai.vercel.app/models/smollm2-1.7b/)2.2 GB at Q4 | Fits541–767 tok/s | Fits555–786 tok/s | Fits1036–1468 tok/s | | [SmolLM-1.7B](https://whatcanirunai.vercel.app/models/smollm-1.7b/)2.2 GB at Q4 | Fits541–767 tok/s | Fits555–786 tok/s | Fits1036–1468 tok/s | | [Spark-X2.6-1.7B](https://whatcanirunai.vercel.app/models/spark-x2.6-1.7b/)2.2 GB at Q4 | Fits542–768 tok/s | Fits556–788 tok/s | Fits1039–1471 tok/s | | [Falcon3-1B-Instruct](https://whatcanirunai.vercel.app/models/falcon3-1b-instruct/)2.2 GB at Q4 | Fits555–786 tok/s | Fits569–806 tok/s | Fits1062–1505 tok/s | | [EuroLLM-1.7B-Instruct](https://whatcanirunai.vercel.app/models/eurollm-1.7b-instruct/)2.2 GB at Q4 | Fits559–792 tok/s | Fits573–812 tok/s | Fits1070–1516 tok/s | | [granite-4.0-1b](https://whatcanirunai.vercel.app/models/granite-4.0-1b/)2.1 GB at Q4 | Fits568–804 tok/s | Fits582–825 tok/s | Fits1087–1540 tok/s | | [LFM2.5-VL-1.6B](https://whatcanirunai.vercel.app/models/lfm2.5-vl-1.6b/)2.1 GB at Q4 | Fits580–822 tok/s | Fits595–843 tok/s | Fits1111–1574 tok/s | | [glm-edge-1.5b-chat](https://whatcanirunai.vercel.app/models/glm-edge-1.5b-chat/)2.1 GB at Q4 | Fits581–824 tok/s | Fits596–845 tok/s | Fits1113–1577 tok/s | | [LFM2-VL-1.6B](https://whatcanirunai.vercel.app/models/lfm2-vl-1.6b/)2.1 GB at Q4 | Fits585–828 tok/s | Fits600–849 tok/s | Fits1119–1585 tok/s | | [Falcon-H1-1.5B-Deep-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-1.5b-deep-instruct/)2.1 GB at Q4 | Fits596–844 tok/s | Fits611–866 tok/s | Fits1141–1616 tok/s | | [Falcon-H1-1.5B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-1.5b-instruct/)2.1 GB at Q4 | Fits596–844 tok/s | Fits611–866 tok/s | Fits1141–1616 tok/s | | [MiniCPM-V-4.7-1B](https://whatcanirunai.vercel.app/models/minicpm-v-4.7-1b/)2.1 GB at Q4 | Fits596–844 tok/s | Fits611–866 tok/s | Fits1141–1616 tok/s | | [Qwen2.5-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-1.5b-instruct/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2.5-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-1.5b/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2.5-Coder-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-1.5b-instruct/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2.5-Coder-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-1.5b/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2.5-Math-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-math-1.5b/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-1.5b-instruct/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Qwen2.5-Math-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-math-1.5b-instruct/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [ReaderLM-v2](https://whatcanirunai.vercel.app/models/readerlm-v2/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [reader-lm-1.5b](https://whatcanirunai.vercel.app/models/reader-lm-1.5b/)2.1 GB at Q4 | Fits600–850 tok/s | Fits615–872 tok/s | Fits1149–1628 tok/s | | [Hymba-1.5B-Instruct](https://whatcanirunai.vercel.app/models/hymba-1.5b-instruct/)2.1 GB at Q4 | Fits608–862 tok/s | Fits624–884 tok/s | Fits1165–1650 tok/s | | [OLMo-2-0425-1B](https://whatcanirunai.vercel.app/models/olmo-2-0425-1b/)2.0 GB at Q4 | Fits624–884 tok/s | Fits640–906 tok/s | Fits1194–1692 tok/s | | [OLMo-2-0425-1B-Instruct](https://whatcanirunai.vercel.app/models/olmo-2-0425-1b-instruct/)2.0 GB at Q4 | Fits624–884 tok/s | Fits640–906 tok/s | Fits1194–1692 tok/s | | [Yi-Coder-1.5B-Chat](https://whatcanirunai.vercel.app/models/yi-coder-1.5b-chat/)2.0 GB at Q4 | Fits627–889 tok/s | Fits643–912 tok/s | Fits1201–1702 tok/s | | [Bolmo-1B](https://whatcanirunai.vercel.app/models/bolmo-1b/)2.0 GB at Q4 | Fits631–893 tok/s | Fits647–916 tok/s | Fits1207–1710 tok/s | | [granite-4.0-h-1b](https://whatcanirunai.vercel.app/models/granite-4.0-h-1b/)2.0 GB at Q4 | Fits634–898 tok/s | Fits650–921 tok/s | Fits1213–1719 tok/s | | [granite-3.1-1b-a400m-instruct](https://whatcanirunai.vercel.app/models/granite-3.1-1b-a400m-instruct/)1.9 GB at Q4 · MoE | Fitsup to 3281 tok/s | Fitsup to 3365 tok/s | Fitsup to 6281 tok/s | | [granite-3.0-1b-a400m-instruct](https://whatcanirunai.vercel.app/models/granite-3.0-1b-a400m-instruct/)1.9 GB at Q4 · MoE | Fitsup to 3281 tok/s | Fitsup to 3365 tok/s | Fitsup to 6281 tok/s | | [MiniCPM-V-4.6](https://whatcanirunai.vercel.app/models/minicpm-v-4.6/)1.9 GB at Q4 | Fits712–1009 tok/s | Fits731–1035 tok/s | Fits1364–1932 tok/s | | [MiniCPM-V-4.6-Thinking](https://whatcanirunai.vercel.app/models/minicpm-v-4.6-thinking/)1.9 GB at Q4 | Fits712–1009 tok/s | Fits731–1035 tok/s | Fits1364–1932 tok/s | | [kanana-2-1.3b-instruct](https://whatcanirunai.vercel.app/models/kanana-2-1.3b-instruct/)1.9 GB at Q4 | Fits717–1016 tok/s | Fits736–1042 tok/s | Fits1373–1945 tok/s | | [OLMo-1B-0724-hf](https://whatcanirunai.vercel.app/models/olmo-1b-0724-hf/)1.9 GB at Q4 | Fits724–1025 tok/s | Fits742–1052 tok/s | Fits1386–1963 tok/s | | [EXAONE-4.0-1.2B](https://whatcanirunai.vercel.app/models/exaone-4.0-1.2b/)1.9 GB at Q4 | Fits724–1026 tok/s | Fits743–1052 tok/s | Fits1386–1964 tok/s | | [Llama-3.2-1B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.2-1b-instruct/)1.9 GB at Q4 | Fits750–1062 tok/s | Fits769–1089 tok/s | Fits1435–2033 tok/s | | [Llama-3.2-1B](https://whatcanirunai.vercel.app/models/llama-3.2-1b/)1.9 GB at Q4 | Fits750–1062 tok/s | Fits769–1089 tok/s | Fits1435–2033 tok/s | | [Llama-3.2-1B](https://whatcanirunai.vercel.app/models/llama-3.2-1b-nousresearch/)1.9 GB at Q4 | Fits750–1062 tok/s | Fits769–1089 tok/s | Fits1435–2033 tok/s | | [Zamba2-1.2B-instruct](https://whatcanirunai.vercel.app/models/zamba2-1.2b-instruct/)1.8 GB at Q4 | Fits762–1080 tok/s | Fits782–1108 tok/s | Fits1460–2068 tok/s | | [HRM-Text-1B](https://whatcanirunai.vercel.app/models/hrm-text-1b/)1.8 GB at Q4 | Fits783–1110 tok/s | Fits803–1138 tok/s | Fits1499–2124 tok/s | | [AMD-OLMo-1B](https://whatcanirunai.vercel.app/models/amd-olmo-1b/)1.8 GB at Q4 | Fits787–1115 tok/s | Fits807–1144 tok/s | Fits1507–2135 tok/s | | [LFM2.5-1.2B-Instruct](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-instruct/)1.8 GB at Q4 | Fits792–1121 tok/s | Fits812–1150 tok/s | Fits1515–2147 tok/s | | [LFM2-1.2B](https://whatcanirunai.vercel.app/models/lfm2-1.2b/)1.8 GB at Q4 | Fits792–1121 tok/s | Fits812–1150 tok/s | Fits1515–2147 tok/s | | [LFM2.5-1.2B-Thinking](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-thinking/)1.8 GB at Q4 | Fits792–1121 tok/s | Fits812–1150 tok/s | Fits1515–2147 tok/s | | [LFM2.5-1.2B-JP-202606](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-jp-202606/)1.8 GB at Q4 | Fits792–1121 tok/s | Fits812–1150 tok/s | Fits1515–2147 tok/s | | [MiniCPM5-1B](https://whatcanirunai.vercel.app/models/minicpm5-1b/)1.8 GB at Q4 | Fits857–1214 tok/s | Fits879–1246 tok/s | Fits1641–2325 tok/s | | [MiniCPM5-1B-SFT](https://whatcanirunai.vercel.app/models/minicpm5-1b-sft/)1.8 GB at Q4 | Fits857–1214 tok/s | Fits879–1246 tok/s | Fits1641–2325 tok/s | | [gemma-3-1b-it](https://whatcanirunai.vercel.app/models/gemma-3-1b-it/)1.7 GB at Q4 | Fits926–1312 tok/s | Fits950–1346 tok/s | Fits1774–2513 tok/s | | [gemma-3-1b-pt](https://whatcanirunai.vercel.app/models/gemma-3-1b-pt/)1.7 GB at Q4 | Fits926–1312 tok/s | Fits950–1346 tok/s | Fits1774–2513 tok/s | | [Qwen3.5-0.8B](https://whatcanirunai.vercel.app/models/qwen3.5-0.8b/)1.6 GB at Q4 | Fits1061–1502 tok/s | Fits1088–1541 tok/s | Fits2031–2877 tok/s | | [Falcon-E-3B-Instruct](https://whatcanirunai.vercel.app/models/falcon-e-3b-instruct/)1.6 GB at Q4 | Fits1072–1519 tok/s | Fits1099–1558 tok/s | Fits2052–2907 tok/s | | [bitnet-b1.58-2B-4T](https://whatcanirunai.vercel.app/models/bitnet-b1.58-2b-4t/)1.6 GB at Q4 | Fits1090–1544 tok/s | Fits1118–1584 tok/s | Fits2087–2957 tok/s | | [FastVLM-0.5B](https://whatcanirunai.vercel.app/models/fastvlm-0.5b/)1.5 GB at Q4 | Fits1221–1729 tok/s | Fits1252–1774 tok/s | Fits2337–3311 tok/s | | [Qwen3-0.6B](https://whatcanirunai.vercel.app/models/qwen3-0.6b/)1.5 GB at Q4 | Fits1232–1746 tok/s | Fits1264–1791 tok/s | Fits2360–3343 tok/s | | [LFM2-700M](https://whatcanirunai.vercel.app/models/lfm2-700m/)1.5 GB at Q4 | Fits1248–1767 tok/s | Fits1280–1813 tok/s | Fits2389–3384 tok/s | | [Falcon-H1-Tiny-R-0.6B](https://whatcanirunai.vercel.app/models/falcon-h1-tiny-r-0.6b/)1.4 GB at Q4 | Fits1489–2109 tok/s | Fits1527–2163 tok/s | Fits2850–4037 tok/s | | [Qwen3-0.6B-Base](https://whatcanirunai.vercel.app/models/qwen3-0.6b-base/)1.4 GB at Q4 | Fits1554–2202 tok/s | Fits1594–2258 tok/s | Fits2975–4215 tok/s | | [d1-omni-600M](https://whatcanirunai.vercel.app/models/d1-omni-600m/)1.4 GB at Q4 | Fits1578–2235 tok/s | Fits1618–2292 tok/s | Fits3021–4279 tok/s | | [HyperCLOVAX-SEED-Text-Instruct-0.5B](https://whatcanirunai.vercel.app/models/hyperclovax-seed-text-instruct-0.5b/)1.4 GB at Q4 | Fits1636–2317 tok/s | Fits1678–2377 tok/s | Fits3132–4437 tok/s | | [Falcon-H1-0.5B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-0.5b-instruct/)1.4 GB at Q4 | Fits1777–2517 tok/s | Fits1822–2581 tok/s | Fits3401–4819 tok/s | | [Qwen2.5-0.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-0.5b-instruct/)1.3 GB at Q4 | Fits1875–2656 tok/s | Fits1923–2724 tok/s | Fits3590–5086 tok/s | | [Qwen2.5-0.5B](https://whatcanirunai.vercel.app/models/qwen2.5-0.5b/)1.3 GB at Q4 | Fits1875–2656 tok/s | Fits1923–2724 tok/s | Fits3590–5086 tok/s | | [Qwen2-0.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-0.5b-instruct/)1.3 GB at Q4 | Fits1875–2656 tok/s | Fits1923–2724 tok/s | Fits3590–5086 tok/s | | [Qwen2.5-Coder-0.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-0.5b-instruct/)1.3 GB at Q4 | Fits1875–2656 tok/s | Fits1923–2724 tok/s | Fits3590–5086 tok/s | | [LFM2-VL-450M](https://whatcanirunai.vercel.app/models/lfm2-vl-450m/)1.3 GB at Q4 | Fits2055–2911 tok/s | Fits2107–2986 tok/s | Fits3934–5573 tok/s | | [LFM2.5-VL-450M](https://whatcanirunai.vercel.app/models/lfm2.5-vl-450m/)1.3 GB at Q4 | Fits2064–2925 tok/s | Fits2117–3000 tok/s | Fits3952–5599 tok/s | | [Apertus-v1.1-0.5B](https://whatcanirunai.vercel.app/models/apertus-v1.1-0.5b/)1.3 GB at Q4 | Fits2113–2994 tok/s | Fits2167–3071 tok/s | Fits4046–5732 tok/s | | [MiniCPM4-0.5B](https://whatcanirunai.vercel.app/models/minicpm4-0.5b/)1.3 GB at Q4 | Fits2135–3025 tok/s | Fits2190–3102 tok/s | Fits4088–5791 tok/s | | [circuit-sparsity](https://whatcanirunai.vercel.app/models/circuit-sparsity/)1.3 GB at Q4 | Fits2210–3131 tok/s | Fits2267–3211 tok/s | Fits4231–5995 tok/s | | [SmolLM2-360M](https://whatcanirunai.vercel.app/models/smollm2-360m/)1.3 GB at Q4 | Fits2560–3627 tok/s | Fits2626–3720 tok/s | Fits4902–6944 tok/s | | [SmolLM-360M](https://whatcanirunai.vercel.app/models/smollm-360m/)1.3 GB at Q4 | Fits2560–3627 tok/s | Fits2626–3720 tok/s | Fits4902–6944 tok/s | | [ERNIE-4.5-0.3B-PT](https://whatcanirunai.vercel.app/models/ernie-4.5-0.3b-pt/)1.3 GB at Q4 | Fits2568–3638 tok/s | Fits2634–3731 tok/s | Fits4916–6965 tok/s | | [LFM2.5-350M](https://whatcanirunai.vercel.app/models/lfm2.5-350m/)1.2 GB at Q4 | Fits2613–3702 tok/s | Fits2680–3797 tok/s | Fits5003–7088 tok/s | | [LFM2-350M](https://whatcanirunai.vercel.app/models/lfm2-350m/)1.2 GB at Q4 | Fits2613–3702 tok/s | Fits2680–3797 tok/s | Fits5003–7088 tok/s | | [granite-4.0-350m](https://whatcanirunai.vercel.app/models/granite-4.0-350m/)1.2 GB at Q4 | Fits2629–3724 tok/s | Fits2696–3820 tok/s | Fits5033–7130 tok/s | | [granite-4.0-h-350m](https://whatcanirunai.vercel.app/models/granite-4.0-h-350m/)1.2 GB at Q4 | Fits2722–3856 tok/s | Fits2792–3955 tok/s | Fits5211–7382 tok/s | Change a part? Open any build in the [planner](https://whatcanirunai.vercel.app/planner/), tweak it, use "Compare this build", and it lands here as Build A. All [build recipes](https://whatcanirunai.vercel.app/builds/). --- Agent guide: https://whatcanirunai.vercel.app/llms.txt · JSON API: https://whatcanirunai.vercel.app/developers/