# How much VRAM do you need for a 1B–2B model?

> A 1B–2B LLM needs about 1.9 GB of VRAM at Q4, 2.5 GB at Q8 and 3.8 GB at FP16. Context-length table and the cheapest GPU and Mac that fit.

Source: https://whatcanirunai.vercel.app/how-much-vram/1b/

# How much VRAM do you need for a 1B–2B model?

Short answer: about **1.9 GB** at Q4\_K\_M, the usual quant for local use. Below: every quant, longer contexts, the cheapest hardware that fits, and every 1B–2B open model we track.

Q4\_K\_M

1.9 GB 0.7 GB weights

Q8\_0

2.5 GB 1.3 GB weights

FP16

3.8 GB 2.5 GB weights

Worked out for [Llama-3.2-1B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.2-1b-instruct/) (1.2B parameters, from the Hugging Face API). Formula: weights × 1.15 + 1 GB at a \~4k context ([how we calculate](https://whatcanirunai.vercel.app/method/)).

Cheapest GPU that fits at Q4: **Intel Arc B570** (10 GB, $219). Cheapest Mac or mini PC: **Mac mini M4 16GB** ($789).

## Cheapest hardware for each quant

Only devices with a sourced price. GPUs at list price (MSRP); street prices may be higher.

__Cheapest GPU and Mac or mini PC for a 1B–2B model at each quant__
| Setting                | Memory needed | Cheapest GPU                                              | Cheapest Mac / mini PC                                            |
| ---------------------- | ------------- | --------------------------------------------------------- | ----------------------------------------------------------------- |
| Q4\_K\_M, \~4k context | 1.9 GB        | [Intel Arc B570](https://whatcanirunai.vercel.app/hardware/arc-b570/) 10 GB usable · $219 | [Mac mini M4 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m4-16/) 12 GB usable · $789 |
| Q8\_0, \~4k context    | 2.5 GB        | [Intel Arc B570](https://whatcanirunai.vercel.app/hardware/arc-b570/) 10 GB usable · $219 | [Mac mini M4 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m4-16/) 12 GB usable · $789 |
| FP16, \~4k context     | 3.8 GB        | [Intel Arc B570](https://whatcanirunai.vercel.app/hardware/arc-b570/) 10 GB usable · $219 | [Mac mini M4 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m4-16/) 12 GB usable · $789 |
| Q4\_K\_M, 32k context  | 2.8 GB        | [Intel Arc B570](https://whatcanirunai.vercel.app/hardware/arc-b570/) 10 GB usable · $219 | [Mac mini M4 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m4-16/) 12 GB usable · $789 |

## Memory by context length

Each extra token of context adds to the KV cache. Numbers for Llama-3.2-1B-Instruct, from its published config; a Q8 KV cache halves the cache part.

__Memory needed for a 1B–2B model by context length__
| Context     | Q4\_K\_M, FP16 cache | Q4\_K\_M, Q8 cache | Q8\_0, FP16 cache |
| ----------- | -------------------- | ------------------ | ----------------- |
| 4k tokens   | 1.9 GB               | 1.9 GB             | 2.5 GB            |
| 8k tokens   | 2.0 GB               | 1.9 GB             | 2.6 GB            |
| 16k tokens  | 2.3 GB               | 2.0 GB             | 2.9 GB            |
| 32k tokens  | 2.8 GB               | 2.3 GB             | 3.4 GB            |
| 64k tokens  | 3.9 GB               | 2.8 GB             | 4.5 GB            |
| 128k tokens | 6.0 GB               | 3.9 GB             | 6.7 GB            |

## Hardware that runs it

Bar: Q4 memory needed vs. usable memory. Speed is estimated decode at Q4 ([how we estimate](https://whatcanirunai.vercel.app/method/#speed)).FitsTight

__Tracked hardware that runs a 1B–2B model at Q4__
| Hardware                                                                                               | Q4 need vs. usable | Q4 / Q8 / FP16       | Est. tok/s |
| ------------------------------------------------------------------------------------------------------ | ------------------ | -------------------- | ---------- |
| [GeForce RTX 2060 6GB](https://whatcanirunai.vercel.app/hardware/rtx-2060/)GPU · $349                                                  | **1.9** / 6.0 GB   | Q4FitsQ8FitsFP16Fits | 269–381    |
| [GeForce RTX 3050 8GB](https://whatcanirunai.vercel.app/hardware/rtx-3050-8gb/)GPU · $249                                              | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 179–254    |
| [Radeon RX 7600](https://whatcanirunai.vercel.app/hardware/rx-7600/)GPU · $269                                                         | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 231–327    |
| [Intel Arc A750](https://whatcanirunai.vercel.app/hardware/arc-a750/)GPU · $289                                                        | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 410–581    |
| [GeForce RTX 4060](https://whatcanirunai.vercel.app/hardware/rtx-4060/)GPU · $299                                                      | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 218–309    |
| [GeForce RTX 5060](https://whatcanirunai.vercel.app/hardware/rtx-5060/)GPU · $299                                                      | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [Radeon RX 9060 XT 8GB](https://whatcanirunai.vercel.app/hardware/rx-9060-xt-8gb/)GPU · $299                                           | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 256–363    |
| [GeForce RTX 5060 Ti 8GB](https://whatcanirunai.vercel.app/hardware/rtx-5060-ti-8gb/)GPU · $379                                        | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 3060 Ti](https://whatcanirunai.vercel.app/hardware/rtx-3060-ti/)GPU · $399                                                | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 2060 Super](https://whatcanirunai.vercel.app/hardware/rtx-2060-super/)GPU · $399                                          | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 4060 Ti 8GB](https://whatcanirunai.vercel.app/hardware/rtx-4060-ti-8gb/)GPU · $399                                        | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 231–327    |
| [GeForce RTX 3070](https://whatcanirunai.vercel.app/hardware/rtx-3070/)GPU · $499                                                      | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 2070](https://whatcanirunai.vercel.app/hardware/rtx-2070/)GPU · $499                                                      | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 2070 Super](https://whatcanirunai.vercel.app/hardware/rtx-2070-super/)GPU · $499                                          | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 3070 Ti](https://whatcanirunai.vercel.app/hardware/rtx-3070-ti/)GPU · $599                                                | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 487–690    |
| [GeForce RTX 2080](https://whatcanirunai.vercel.app/hardware/rtx-2080/)GPU · $699                                                      | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 2080 Super](https://whatcanirunai.vercel.app/hardware/rtx-2080-super/)GPU · $699                                          | **1.9** / 8.0 GB   | Q4FitsQ8FitsFP16Fits | 397–563    |
| [Intel Arc B570](https://whatcanirunai.vercel.app/hardware/arc-b570/)GPU · $219                                                        | **1.9** / 10.0 GB  | Q4FitsQ8FitsFP16Fits | 304–431    |
| [GeForce RTX 3080 10GB](https://whatcanirunai.vercel.app/hardware/rtx-3080-10gb/)GPU · $699                                            | **1.9** / 10.0 GB  | Q4FitsQ8FitsFP16Fits | 609–862    |
| [GeForce RTX 2080 Ti](https://whatcanirunai.vercel.app/hardware/rtx-2080-ti/)GPU · $999                                                | **1.9** / 11.0 GB  | Q4FitsQ8FitsFP16Fits | 493–699    |
| [GeForce RTX 2060 12GB](https://whatcanirunai.vercel.app/hardware/rtx-2060-12gb/)GPU                                                   | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 269–381    |
| [Intel Arc B580](https://whatcanirunai.vercel.app/hardware/arc-b580/)GPU · $249                                                        | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 365–517    |
| [GeForce RTX 3060 12GB](https://whatcanirunai.vercel.app/hardware/rtx-3060-12gb/)GPU · $329                                            | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 288–408    |
| [Radeon RX 7700 XT](https://whatcanirunai.vercel.app/hardware/rx-7700-xt/)GPU · $449                                                   | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 346–490    |
| [Radeon RX 6700 XT](https://whatcanirunai.vercel.app/hardware/rx-6700-xt/)GPU · $479                                                   | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 308–436    |
| [GeForce RTX 5070](https://whatcanirunai.vercel.app/hardware/rtx-5070/)GPU · $549                                                      | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 538–762    |
| [Radeon RX 6750 XT](https://whatcanirunai.vercel.app/hardware/rx-6750-xt/)GPU · $549                                                   | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 346–490    |
| [GeForce RTX 4070](https://whatcanirunai.vercel.app/hardware/rtx-4070/)GPU · $599                                                      | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 404–572    |
| [GeForce RTX 4070 Super](https://whatcanirunai.vercel.app/hardware/rtx-4070-super/)GPU · $599                                          | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 404–572    |
| [Mac mini M4 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m4-16/)unified memory · $789                                     | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 72–128     |
| [GeForce RTX 3080 12GB](https://whatcanirunai.vercel.app/hardware/rtx-3080-12gb/)GPU · $799                                            | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 730–1035   |
| [GeForce RTX 4070 Ti](https://whatcanirunai.vercel.app/hardware/rtx-4070-ti/)GPU · $799                                                | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 404–572    |
| [Mac mini M6 16GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m6-16/)unified memory · $899                                     | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 92–163     |
| [GeForce RTX 3080 Ti](https://whatcanirunai.vercel.app/hardware/rtx-3080-ti/)GPU · $1,199                                              | **1.9** / 12.0 GB  | Q4FitsQ8FitsFP16Fits | 730–1035   |
| [Radeon RX 7600 XT](https://whatcanirunai.vercel.app/hardware/rx-7600-xt/)GPU · $329                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 231–327    |
| [Radeon RX 9060 XT 16GB](https://whatcanirunai.vercel.app/hardware/rx-9060-xt-16gb/)GPU · $349                                         | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 256–363    |
| [Intel Arc A770 16GB](https://whatcanirunai.vercel.app/hardware/arc-a770-16gb/)GPU · $349                                              | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 448–635    |
| [GeForce RTX 5060 Ti 16GB](https://whatcanirunai.vercel.app/hardware/rtx-5060-ti-16gb/)GPU · $429                                      | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 359–508    |
| [GeForce RTX 4060 Ti 16GB](https://whatcanirunai.vercel.app/hardware/rtx-4060-ti-16gb/)GPU · $499                                      | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 231–327    |
| [Radeon RX 7800 XT](https://whatcanirunai.vercel.app/hardware/rx-7800-xt/)GPU · $499                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 500–708    |
| [Radeon RX 7900 GRE](https://whatcanirunai.vercel.app/hardware/rx-7900-gre/)GPU · $549                                                 | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 461–653    |
| [Radeon RX 9070](https://whatcanirunai.vercel.app/hardware/rx-9070/)GPU · $549                                                         | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 513–726    |
| [Radeon RX 6800](https://whatcanirunai.vercel.app/hardware/rx-6800/)GPU · $579                                                         | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 410–581    |
| [Radeon RX 9070 XT](https://whatcanirunai.vercel.app/hardware/rx-9070-xt/)GPU · $599                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 513–726    |
| [Radeon RX 6800 XT](https://whatcanirunai.vercel.app/hardware/rx-6800-xt/)GPU · $649                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 410–581    |
| [NVIDIA RTX 2000 Ada 16GB](https://whatcanirunai.vercel.app/hardware/rtx-2000-ada/)GPU · $649                                          | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 179–254    |
| [GeForce RTX 5070 Ti](https://whatcanirunai.vercel.app/hardware/rtx-5070-ti/)GPU · $749                                                | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 718–1017   |
| [GeForce RTX 4070 Ti Super](https://whatcanirunai.vercel.app/hardware/rtx-4070-ti-super/)GPU · $799                                    | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 538–762    |
| [GeForce RTX 4080 Super](https://whatcanirunai.vercel.app/hardware/rtx-4080-super/)GPU · $999                                          | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 589–835    |
| [GeForce RTX 5080](https://whatcanirunai.vercel.app/hardware/rtx-5080/)GPU · $999                                                      | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 769–1089   |
| [Radeon RX 6900 XT](https://whatcanirunai.vercel.app/hardware/rx-6900-xt/)GPU · $999                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 410–581    |
| [NVIDIA RTX A4000](https://whatcanirunai.vercel.app/hardware/rtx-a4000/)GPU · $1,000                                                   | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 359–508    |
| [Radeon RX 6950 XT](https://whatcanirunai.vercel.app/hardware/rx-6950-xt/)GPU · $1,099                                                 | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 461–653    |
| [GeForce RTX 4080](https://whatcanirunai.vercel.app/hardware/rtx-4080/)GPU · $1,199                                                    | **1.9** / 16.0 GB  | Q4FitsQ8FitsFP16Fits | 574–813    |
| [Mac mini M6 24GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m6-24/)unified memory · $1,299                                   | **1.9** / 18.0 GB  | Q4FitsQ8FitsFP16Fits | 102–182    |
| [Radeon RX 7900 XT](https://whatcanirunai.vercel.app/hardware/rx-7900-xt/)GPU · $899                                                   | **1.9** / 20.0 GB  | Q4FitsQ8FitsFP16Fits | 641–908    |
| [NVIDIA RTX 4000 Ada 20GB](https://whatcanirunai.vercel.app/hardware/rtx-4000-ada/)GPU · $1,250                                        | **1.9** / 20.0 GB  | Q4FitsQ8FitsFP16Fits | 288–408    |
| [NVIDIA Tesla P40 24GB](https://whatcanirunai.vercel.app/hardware/tesla-p40/)GPU                                                       | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 277–393    |
| [Radeon RX 7900 XTX](https://whatcanirunai.vercel.app/hardware/rx-7900-xtx/)GPU · $999                                                 | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 769–1089   |
| [GeForce RTX 3090](https://whatcanirunai.vercel.app/hardware/rtx-3090/)GPU · $1,499                                                    | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 750–1062   |
| [GeForce RTX 4090](https://whatcanirunai.vercel.app/hardware/rtx-4090/)GPU · $1,599                                                    | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 807–1144   |
| [Mac mini M6 32GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m6-32/)unified memory · $1,799                                   | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 102–182    |
| [GeForce RTX 3090 Ti](https://whatcanirunai.vercel.app/hardware/rtx-3090-ti/)GPU · $1,999                                              | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 807–1144   |
| [NVIDIA RTX A5000](https://whatcanirunai.vercel.app/hardware/rtx-a5000/)GPU · $2,250                                                   | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 615–871    |
| [MacBook Pro M6 32GB](https://whatcanirunai.vercel.app/hardware/macbook-pro-m6-32/)unified memory · $2,399                             | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 102–182    |
| [NVIDIA Titan RTX](https://whatcanirunai.vercel.app/hardware/titan-rtx/)GPU · $2,499                                                   | **1.9** / 24.0 GB  | Q4FitsQ8FitsFP16Fits | 538–762    |
| [NVIDIA Tesla V100 32GB (PCIe)](https://whatcanirunai.vercel.app/hardware/tesla-v100-32gb/)GPU                                         | **1.9** / 32.0 GB  | Q4FitsQ8FitsFP16Fits | 721–1021   |
| [AMD Radeon Instinct MI50 32GB](https://whatcanirunai.vercel.app/hardware/instinct-mi50-32gb/)GPU                                      | **1.9** / 32.0 GB  | Q4FitsQ8FitsFP16Fits | 820–1162   |
| [AMD Radeon Instinct MI60 32GB](https://whatcanirunai.vercel.app/hardware/instinct-mi60/)GPU                                           | **1.9** / 32.0 GB  | Q4FitsQ8FitsFP16Fits | 820–1162   |
| [Intel Arc Pro B70 32GB](https://whatcanirunai.vercel.app/hardware/arc-pro-b70/)GPU · $949                                             | **1.9** / 32.0 GB  | Q4FitsQ8FitsFP16Fits | 487–690    |
| [GeForce RTX 5090](https://whatcanirunai.vercel.app/hardware/rtx-5090/)GPU · $1,999                                                    | **1.9** / 32.0 GB  | Q4FitsQ8FitsFP16Fits | 1435–2033  |
| [Mac mini M5 Pro 64GB](https://whatcanirunai.vercel.app/hardware/mac-mini-m5-pro-64/)unified memory · $3,199                           | **1.9** / 48.0 GB  | Q4FitsQ8FitsFP16Fits | 184–328    |
| [NVIDIA RTX A6000](https://whatcanirunai.vercel.app/hardware/rtx-a6000/)GPU · $4,649                                                   | **1.9** / 48.0 GB  | Q4FitsQ8FitsFP16Fits | 615–871    |
| [NVIDIA DGX Spark 64GB](https://whatcanirunai.vercel.app/hardware/dgx-spark-64/)unified memory · $4,999                                | **1.9** / 48.0 GB  | Q4FitsQ8FitsFP16Fits | 164–292    |
| [NVIDIA RTX 6000 Ada 48GB](https://whatcanirunai.vercel.app/hardware/rtx-6000-ada/)GPU · $6,799                                        | **1.9** / 48.0 GB  | Q4FitsQ8FitsFP16Fits | 769–1089   |
| [Framework Desktop (Ryzen AI Max+ 395) 128GB](https://whatcanirunai.vercel.app/hardware/framework-desktop-128/)unified memory · $3,449 | **1.9** / 96.0 GB  | Q4FitsQ8FitsFP16Fits | 154–273    |
| [MacBook Pro M4 Max 128GB](https://whatcanirunai.vercel.app/hardware/macbook-pro-m4-max-128/)unified memory · $4,999                   | **1.9** / 96.0 GB  | Q4FitsQ8FitsFP16Fits | 328–583    |
| [MacBook Pro M5 Max 128GB](https://whatcanirunai.vercel.app/hardware/macbook-pro-m5-max-128/)unified memory · $6,699                   | **1.9** / 96.0 GB  | Q4FitsQ8FitsFP16Fits | 369–656    |
| [NVIDIA DGX Spark 128GB](https://whatcanirunai.vercel.app/hardware/dgx-spark-128/)unified memory · $6,950                              | **1.9** / 96.0 GB  | Q4FitsQ8FitsFP16Fits | 164–292    |
| [NVIDIA RTX Pro 6000 Blackwell 96GB](https://whatcanirunai.vercel.app/hardware/rtx-pro-6000-blackwell/)GPU · $8,565                    | **1.9** / 96.0 GB  | Q4FitsQ8FitsFP16Fits | 1435–2033  |
| [Mac Studio M5 Ultra 256GB](https://whatcanirunai.vercel.app/hardware/mac-studio-m5-ultra-256/)unified memory · $9,499                 | **1.9** / 192.0 GB | Q4FitsQ8FitsFP16Fits | 721–1281   |
| [Mac Studio M5 Ultra 512GB](https://whatcanirunai.vercel.app/hardware/mac-studio-m5-ultra-512/)unified memory · $22,299                | **1.9** / 384.0 GB | Q4FitsQ8FitsFP16Fits | 721–1281   |

## 1B–2B models we track

__Memory needed by each 1B–2B model we track__
| Model                                                                                                  | Q4\_K\_M | Q8\_0  | FP16   |
| ------------------------------------------------------------------------------------------------------ | -------- | ------ | ------ |
| [MiniCPM5-1B](https://whatcanirunai.vercel.app/models/minicpm5-1b/)1.1B params · openbmb                                               | 1.8 GB   | 2.3 GB | 3.5 GB |
| [MiniCPM5-1B-SFT](https://whatcanirunai.vercel.app/models/minicpm5-1b-sft/)1.1B params · openbmb                                       | 1.8 GB   | 2.3 GB | 3.5 GB |
| [LFM2.5-1.2B-Instruct](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-instruct/)1.2B params · LiquidAI                            | 1.8 GB   | 2.4 GB | 3.7 GB |
| [LFM2-1.2B](https://whatcanirunai.vercel.app/models/lfm2-1.2b/)1.2B params · LiquidAI                                                  | 1.8 GB   | 2.4 GB | 3.7 GB |
| [LFM2.5-1.2B-Thinking](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-thinking/)1.2B params · LiquidAI                            | 1.8 GB   | 2.4 GB | 3.7 GB |
| [LFM2.5-1.2B-JP-202606](https://whatcanirunai.vercel.app/models/lfm2.5-1.2b-jp-202606/)1.2B params · LiquidAI                          | 1.8 GB   | 2.4 GB | 3.7 GB |
| [AMD-OLMo-1B](https://whatcanirunai.vercel.app/models/amd-olmo-1b/)1.2B params · amd                                                   | 1.8 GB   | 2.4 GB | 3.7 GB |
| [HRM-Text-1B](https://whatcanirunai.vercel.app/models/hrm-text-1b/)1.2B params · sapientinc                                            | 1.8 GB   | 2.4 GB | 3.7 GB |
| [Zamba2-1.2B-instruct](https://whatcanirunai.vercel.app/models/zamba2-1.2b-instruct/)1.2B params · Zyphra                              | 1.8 GB   | 2.5 GB | 3.8 GB |
| [Llama-3.2-1B-Instruct](https://whatcanirunai.vercel.app/models/llama-3.2-1b-instruct/)1.2B params · meta-llama                        | 1.9 GB   | 2.5 GB | 3.8 GB |
| [Llama-3.2-1B](https://whatcanirunai.vercel.app/models/llama-3.2-1b/)1.2B params · meta-llama                                          | 1.9 GB   | 2.5 GB | 3.8 GB |
| [Llama-3.2-1B](https://whatcanirunai.vercel.app/models/llama-3.2-1b-nousresearch/)1.2B params · NousResearch                           | 1.9 GB   | 2.5 GB | 3.8 GB |
| [EXAONE-4.0-1.2B](https://whatcanirunai.vercel.app/models/exaone-4.0-1.2b/)1.3B params · LGAI-EXAONE                                   | 1.9 GB   | 2.6 GB | 3.9 GB |
| [OLMo-1B-0724-hf](https://whatcanirunai.vercel.app/models/olmo-1b-0724-hf/)1.3B params · allenai                                       | 1.9 GB   | 2.6 GB | 3.9 GB |
| [kanana-2-1.3b-instruct](https://whatcanirunai.vercel.app/models/kanana-2-1.3b-instruct/)1.3B params · kakaocorp                       | 1.9 GB   | 2.6 GB | 4.0 GB |
| [MiniCPM-V-4.6](https://whatcanirunai.vercel.app/models/minicpm-v-4.6/)1.3B params · openbmb                                           | 1.9 GB   | 2.6 GB | 4.0 GB |
| [MiniCPM-V-4.6-Thinking](https://whatcanirunai.vercel.app/models/minicpm-v-4.6-thinking/)1.3B params · openbmb                         | 1.9 GB   | 2.6 GB | 4.0 GB |
| [granite-3.1-1b-a400m-instruct](https://whatcanirunai.vercel.app/models/granite-3.1-1b-a400m-instruct/)1.3B params · ibm-granite · MoE | 1.9 GB   | 2.6 GB | 4.1 GB |
| [granite-3.0-1b-a400m-instruct](https://whatcanirunai.vercel.app/models/granite-3.0-1b-a400m-instruct/)1.3B params · ibm-granite · MoE | 1.9 GB   | 2.6 GB | 4.1 GB |
| [granite-4.0-h-1b](https://whatcanirunai.vercel.app/models/granite-4.0-h-1b/)1.5B params · ibm-granite                                 | 2.0 GB   | 2.8 GB | 4.4 GB |
| [Bolmo-1B](https://whatcanirunai.vercel.app/models/bolmo-1b/)1.5B params · allenai                                                     | 2.0 GB   | 2.8 GB | 4.4 GB |
| [Yi-Coder-1.5B-Chat](https://whatcanirunai.vercel.app/models/yi-coder-1.5b-chat/)1.5B params · 01-ai                                   | 2.0 GB   | 2.8 GB | 4.4 GB |
| [OLMo-2-0425-1B](https://whatcanirunai.vercel.app/models/olmo-2-0425-1b/)1.5B params · allenai                                         | 2.0 GB   | 2.8 GB | 4.4 GB |
| [OLMo-2-0425-1B-Instruct](https://whatcanirunai.vercel.app/models/olmo-2-0425-1b-instruct/)1.5B params · allenai                       | 2.0 GB   | 2.8 GB | 4.4 GB |
| [Hymba-1.5B-Instruct](https://whatcanirunai.vercel.app/models/hymba-1.5b-instruct/)1.5B params · nvidia                                | 2.1 GB   | 2.9 GB | 4.5 GB |
| [Qwen2.5-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-1.5b-instruct/)1.5B params · Qwen                              | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2.5-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-1.5b/)1.5B params · Qwen                                                | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2.5-Coder-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-coder-1.5b-instruct/)1.5B params · Qwen                  | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2.5-Coder-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-coder-1.5b/)1.5B params · Qwen                                    | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2.5-Math-1.5B](https://whatcanirunai.vercel.app/models/qwen2.5-math-1.5b/)1.5B params · Qwen                                      | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2-1.5b-instruct/)1.5B params · Qwen                                  | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Qwen2.5-Math-1.5B-Instruct](https://whatcanirunai.vercel.app/models/qwen2.5-math-1.5b-instruct/)1.5B params · Qwen                    | 2.1 GB   | 2.9 GB | 4.6 GB |
| [ReaderLM-v2](https://whatcanirunai.vercel.app/models/readerlm-v2/)1.5B params · jinaai                                                | 2.1 GB   | 2.9 GB | 4.6 GB |
| [reader-lm-1.5b](https://whatcanirunai.vercel.app/models/reader-lm-1.5b/)1.5B params · jinaai                                          | 2.1 GB   | 2.9 GB | 4.6 GB |
| [MiniCPM-V-4.7-1B](https://whatcanirunai.vercel.app/models/minicpm-v-4.7-1b/)1.6B params · openbmb                                     | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Falcon-H1-1.5B-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-1.5b-instruct/)1.6B params · tiiuae                        | 2.1 GB   | 2.9 GB | 4.6 GB |
| [Falcon-H1-1.5B-Deep-Instruct](https://whatcanirunai.vercel.app/models/falcon-h1-1.5b-deep-instruct/)1.6B params · tiiuae              | 2.1 GB   | 2.9 GB | 4.6 GB |
| [LFM2-VL-1.6B](https://whatcanirunai.vercel.app/models/lfm2-vl-1.6b/)1.6B params · LiquidAI                                            | 2.1 GB   | 2.9 GB | 4.6 GB |
| [glm-edge-1.5b-chat](https://whatcanirunai.vercel.app/models/glm-edge-1.5b-chat/)1.6B params · zai-org                                 | 2.1 GB   | 2.9 GB | 4.7 GB |
| [LFM2.5-VL-1.6B](https://whatcanirunai.vercel.app/models/lfm2.5-vl-1.6b/)1.6B params · LiquidAI                                        | 2.1 GB   | 3.0 GB | 4.7 GB |
| [granite-4.0-1b](https://whatcanirunai.vercel.app/models/granite-4.0-1b/)1.6B params · ibm-granite                                     | 2.1 GB   | 3.0 GB | 4.8 GB |
| [EuroLLM-1.7B-Instruct](https://whatcanirunai.vercel.app/models/eurollm-1.7b-instruct/)1.7B params · utter-project                     | 2.2 GB   | 3.0 GB | 4.8 GB |
| [Falcon3-1B-Instruct](https://whatcanirunai.vercel.app/models/falcon3-1b-instruct/)1.7B params · tiiuae                                | 2.2 GB   | 3.0 GB | 4.8 GB |
| [Spark-X2.6-1.7B](https://whatcanirunai.vercel.app/models/spark-x2.6-1.7b/)1.7B params · XHToken                                       | 2.2 GB   | 3.1 GB | 4.9 GB |
| [SmolLM2-1.7B](https://whatcanirunai.vercel.app/models/smollm2-1.7b/)1.7B params · HuggingFaceTB                                       | 2.2 GB   | 3.1 GB | 4.9 GB |
| [SmolLM-1.7B](https://whatcanirunai.vercel.app/models/smollm-1.7b/)1.7B params · HuggingFaceTB                                         | 2.2 GB   | 3.1 GB | 4.9 GB |
| [Qwen3-1.7B-Base](https://whatcanirunai.vercel.app/models/qwen3-1.7b-base/)1.7B params · Qwen                                          | 2.2 GB   | 3.1 GB | 5.0 GB |
| [LFM3-1.7B](https://whatcanirunai.vercel.app/models/lfm3-1.7b/)1.7B params · LiquidAI                                                  | 2.2 GB   | 3.1 GB | 5.0 GB |
| [DeepSeek-R1-Distill-Qwen-1.5B](https://whatcanirunai.vercel.app/models/deepseek-r1-distill-qwen-1.5b/)1.8B params · deepseek-ai       | 2.2 GB   | 3.2 GB | 5.1 GB |
| [internlm2\_5-1\_8b-chat](https://whatcanirunai.vercel.app/models/internlm2-5-1-8b-chat/)1.9B params · internlm                        | 2.3 GB   | 3.3 GB | 5.3 GB |
| [FastVLM-1.5B](https://whatcanirunai.vercel.app/models/fastvlm-1.5b/)1.9B params · apple                                               | 2.3 GB   | 3.3 GB | 5.4 GB |
| [LFM3-VL-2B](https://whatcanirunai.vercel.app/models/lfm3-vl-2b/)1.9B params · LiquidAI                                                | 2.3 GB   | 3.3 GB | 5.4 GB |
| [d1-2B](https://whatcanirunai.vercel.app/models/d1-2b/)1.9B params · LiquidAI                                                          | 2.3 GB   | 3.3 GB | 5.4 GB |
| [Youtu-LLM-2B](https://whatcanirunai.vercel.app/models/youtu-llm-2b/)2.0B params · tencent                                             | 2.4 GB   | 3.4 GB | 5.5 GB |

Have a different amount of memory? Try the [VRAM calculator](https://whatcanirunai.vercel.app/vram-calculator/) or see [best LLMs by VRAM size](https://whatcanirunai.vercel.app/best-llm-for/).

## Quick answers

How much VRAM do I need for a 1B–2B model?

About 1.9 GB at Q4\_K\_M, 2.5 GB at Q8\_0 and 3.8 GB at FP16, at a \~4k context. That is for Llama-3.2-1B-Instruct (1.2B parameters): weights plus about 15% for the KV cache and 1 GB for the runtime.

How much VRAM does a 1B–2B model need for 32k context?

About 2.8 GB at Q4\_K\_M with an FP16 KV cache, or 2.3 GB with a Q8 cache, based on Llama-3.2-1B-Instruct's published config. Models with different attention layouts need more or less.

What is the cheapest GPU for a 1B–2B model?

Of the GPUs we track, the Intel Arc B570 (10 GB, $219 list price) is the cheapest that runs it fully on the GPU at Q4.

Can a Mac run a 1B–2B model?

Yes. The cheapest Mac or unified-memory mini PC we track that runs it at Q4 is the Mac mini M4 16GB (16 GB, about 12.0 GB usable, $789).

Does a MoE model of the same size need less VRAM?

No. A mixture-of-experts model still has to hold all of its weights in memory, so it needs about the same VRAM as a dense model with the same total parameters. It runs faster, because only a few experts are read per token.

Other sizes: [2B–3B](https://whatcanirunai.vercel.app/how-much-vram/3b/) · [4B–6B](https://whatcanirunai.vercel.app/how-much-vram/4b/) · [7B–8B](https://whatcanirunai.vercel.app/how-much-vram/8b/) · [12B–14B](https://whatcanirunai.vercel.app/how-much-vram/14b/) · [16B–21B](https://whatcanirunai.vercel.app/how-much-vram/20b/) · [24B–27B](https://whatcanirunai.vercel.app/how-much-vram/27b/) · [30B–32B](https://whatcanirunai.vercel.app/how-much-vram/32b/) · [40B–50B](https://whatcanirunai.vercel.app/how-much-vram/49b/) · [70B](https://whatcanirunai.vercel.app/how-much-vram/70b/) · [120B](https://whatcanirunai.vercel.app/how-much-vram/120b/) · [200B–300B](https://whatcanirunai.vercel.app/how-much-vram/235b/) · [300B–480B](https://whatcanirunai.vercel.app/how-much-vram/400b/) · [500B–800B](https://whatcanirunai.vercel.app/how-much-vram/671b/) · [1T+](https://whatcanirunai.vercel.app/how-much-vram/1t/) · [All sizes](https://whatcanirunai.vercel.app/how-much-vram/) · [Ollama VRAM by tag](https://whatcanirunai.vercel.app/ollama-vram-requirements/) · [GGUF sizes by quant](https://whatcanirunai.vercel.app/gguf-vram-requirements/)

---
Agent guide: https://whatcanirunai.vercel.app/llms.txt · JSON API: https://whatcanirunai.vercel.app/developers/