# Can I run gpt-oss-20b on an RTX 3090? | What Can I Run AI?

> Yes. gpt-oss-20b at Q4_K_M needs about 15.6 GB; RTX 3090 has 24.0 GB usable. Estimated tokens/sec and context headroom.

Source: https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-3090/

[Home](https://whatcanirunai.vercel.app/) / [Can I run it?](https://whatcanirunai.vercel.app/can-i-run/) / [gpt-oss-20b](https://whatcanirunai.vercel.app/models/gpt-oss-20b/) / gpt-oss-20b on an RTX 3090

**Yes.** gpt-oss-20b at Q4\_K\_M needs about 15.6 GB, and the GeForce RTX 3090 has 24.0 GB usable at a \~4k context.

# Can I run gpt-oss-20b on an RTX 3090?

gpt-oss-20b at Q4\_K\_M needs about 15.6 GB, and the GeForce RTX 3090 has 24.0 GB usable. That figure is 24 GB of VRAM on the spec sheet, of which 24.0 GB is usable for weights and cache, at 936 GB/s memory bandwidth. OpenAI’s open 20B-class gpt-oss model. Only a few billion parameters are active per token, so decode can be quicker than a dense 20B. These are estimates from [our formula](https://whatcanirunai.vercel.app/method/), not measured benchmarks.

Quant that fits

Q4\_K\_M Fits with headroom

Memory needed

15.6 GB at Q4\_K\_M, \~4k context

Usable VRAM

24.0 GB of 24 GB VRAM

Est. tokens/sec

up to 365 estimate

[gpt-oss-20b](https://whatcanirunai.vercel.app/models/gpt-oss-20b/) · [GeForce RTX 3090](https://whatcanirunai.vercel.app/hardware/rtx-3090/) · [Everything this machine runs](https://whatcanirunai.vercel.app/what-can-i-run/rtx-3090/) · [Q4 vs Q8 vs FP16](https://whatcanirunai.vercel.app/quantization/)

## Memory at each quant

| Quant    | Need vs 24.0 GB | Verdict       |
| -------- | --------------- | ------------- |
| Q4\_K\_M | 15.6 GB         | Q4Fits        |
| Q8\_0    | 26.6 GB         | Q8Won’t fit   |
| FP16     | 49.1 GB         | FP16Won’t fit |

## Estimated tokens per second

Estimate: **up to 365 tokens/sec** at Q4\_K\_M (upper bound: this is a Mixture-of-Experts, and only the active weights are counted). This is a bandwidth-derived range, not a bench. Efficiency is a 60–85% GPU band from [How we calculate](https://whatcanirunai.vercel.app/method/).

## Context headroom

At Q4\_K\_M, this machine has headroom out to about **128k tokens** under our KV estimate. Longer chats need more memory for the cache. Figures assume an FP16 KV cache.

__Context headroom for gpt-oss-20b on an RTX 3090 at Q4\_K\_M__
| Context | KV cache (FP16) | Total at Q4\_K\_M | Verdict |
| ------- | --------------- | ----------------- | ------- |
| 4k      | 0.1 GB          | 15.6 GB           | Fits    |
| 8k      | 0.2 GB          | 15.7 GB           | Fits    |
| 16k     | 0.4 GB          | 15.9 GB           | Fits    |
| 32k     | 0.8 GB          | 16.3 GB           | Fits    |
| 64k     | 1.6 GB          | 17.1 GB           | Fits    |
| 128k    | 3.2 GB          | 18.7 GB           | Fits    |

From the model’s config. Try other sizes in the [VRAM calculator](https://whatcanirunai.vercel.app/vram-calculator/) or see the [context length guide](https://whatcanirunai.vercel.app/context-length/).

## What to do instead

* Q8\_0 needs 26.6 GB and does not fit here. The [GeForce RTX 5090](https://whatcanirunai.vercel.app/hardware/rtx-5090/) has 32.0 GB usable, enough for Q8.
* See the [best GPU for this model family](https://whatcanirunai.vercel.app/best-gpu-for/gpt-oss/) and the [gpt-oss-20b model page](https://whatcanirunai.vercel.app/models/gpt-oss-20b/).

## Other hardware for gpt-oss-20b

* [gpt-oss-20b on an RTX 3060 12GB](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-3060-12gb/)Won’t fit
* [gpt-oss-20b on an RTX 3070 Ti](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-3070-ti/)Won’t fit
* [gpt-oss-20b on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-3080-10gb/)Won’t fit
* [gpt-oss-20b on an RTX 4060 Ti 16GB](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-4060-ti-16gb/)Tight
* [gpt-oss-20b on an RTX 4070 Ti Super](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-4070-ti-super/)Tight
* [gpt-oss-20b on an RTX 4090](https://whatcanirunai.vercel.app/can-i-run/gpt-oss-20b/rtx-4090/)Fits

## Other models on an RTX 3090

* [Qwen3-32B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/qwen3-32b/rtx-3090/)Tight
* [Qwen3-30B-A3B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/qwen3-30b-a3b/rtx-3090/)Tight
* [Qwen3.8-27B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3090/)Fits
* [Gemma 3 27B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/gemma-3-27b-it/rtx-3090/)Fits
* [Gemma 4 31B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/gemma-4-31b-it/rtx-3090/)Tight
* [Mistral Small 24B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/mistral-small-3.2-24b-instruct-2506/rtx-3090/)Fits

## Quick answers

Can I run gpt-oss-20b on an RTX 3090?

Yes. gpt-oss-20b at Q4\_K\_M needs about 15.6 GB, and the GeForce RTX 3090 has 24.0 GB usable.

How much VRAM does gpt-oss-20b need?

About 15.6 GB at Q4\_K\_M, 26.6 GB at Q8\_0 and 49.1 GB at FP16, estimated for a \~4k context.

How fast is gpt-oss-20b on an RTX 3090?

Our estimate is up to 365 tokens per second at Q4\_K\_M, as an upper bound for this MoE. It is not a measured benchmark.

What context length can I use?

At Q4\_K\_M, our KV estimate supports up to about 128k tokens before the cache pushes the total over usable memory.

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