# Can I run Qwen3.8-27B on an RTX 3080? | What Can I Run AI?

> No. Qwen3.8-27B needs about 20.4 GB at Q4_K_M, and RTX 3080 has 10.0 GB usable. What to run instead.

Source: https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3080-10gb/

[Home](https://whatcanirunai.vercel.app/) / [Can I run it?](https://whatcanirunai.vercel.app/can-i-run/) / [Qwen3.8-27B](https://whatcanirunai.vercel.app/models/qwen3.8-27b/) / Qwen3.8-27B on an RTX 3080

**No.** Qwen3.8-27B at Q4\_K\_M needs about 20.4 GB, and the GeForce RTX 3080 10GB has 10.0 GB usable at a \~4k context.

# Can I run Qwen3.8-27B on an RTX 3080?

Qwen3.8-27B at Q4\_K\_M needs about 20.4 GB, and the GeForce RTX 3080 10GB has 10.0 GB usable. That figure is 10 GB of VRAM on the spec sheet, of which 10.0 GB is usable for weights and cache, at 760 GB/s memory bandwidth. Alibaba’s denser Qwen3.8 27B, a newer mid-size chat model in the same memory class as Gemma 3 27B. These are estimates from [our formula](https://whatcanirunai.vercel.app/method/), not measured benchmarks.

Quant that fits

None of Q4, Q8 or FP16 needs 20.4 GB

Memory needed

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

Usable VRAM

10.0 GB of 10 GB VRAM

Est. tokens/sec

n/a does not fit

[Qwen3.8-27B](https://whatcanirunai.vercel.app/models/qwen3.8-27b/) · [GeForce RTX 3080 10GB](https://whatcanirunai.vercel.app/hardware/rtx-3080-10gb/) · [Everything this machine runs](https://whatcanirunai.vercel.app/what-can-i-run/rtx-3080-10gb/) · [Q4 vs Q8 vs FP16](https://whatcanirunai.vercel.app/quantization/)

## Memory at each quant

| Quant    | Need vs 10.0 GB | Verdict       |
| -------- | --------------- | ------------- |
| Q4\_K\_M | 20.4 GB         | Q4Won’t fit   |
| Q8\_0    | 34.9 GB         | Q8Won’t fit   |
| FP16     | 64.9 GB         | FP16Won’t fit |

## Estimated tokens per second

No speed estimate: Qwen3.8-27B does not fit in the 10.0 GB this machine can use, so we do not publish a tokens/sec figure for CPU offload.

## Context headroom

Even at a short \~4k context, the weights do not fit. Extending context would only increase the gap.

__Context headroom for Qwen3.8-27B on an RTX 3080 at Q4\_K\_M__
| Context | KV cache (FP16) | Total at Q4\_K\_M | Verdict   |
| ------- | --------------- | ----------------- | --------- |
| 4k      | 0.3 GB          | 20.4 GB           | Won’t fit |
| 8k      | 0.5 GB          | 20.6 GB           | Won’t fit |
| 16k     | 1.1 GB          | 21.2 GB           | Won’t fit |
| 32k     | 2.1 GB          | 22.2 GB           | Won’t fit |
| 64k     | 4.3 GB          | 24.4 GB           | Won’t fit |
| 128k    | 8.6 GB          | 28.7 GB           | Won’t fit |

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

* Run [Gemma 3 12B](https://whatcanirunai.vercel.app/can-i-run/gemma-3-12b-it/rtx-3080-10gb/) instead (9.5 GB at Q4). It is a different model, not a drop-in, but it does fit this machine.
* To hold Qwen3.8-27B at Q4, you need about 20.4 GB. The [GeForce RTX 3090](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3090/) has 24.0 GB usable ($1499 list price).
* CPU offload can load the weights anyway, but it is slow. We do not publish an offload tokens/sec figure.
* See the [best GPU for this model family](https://whatcanirunai.vercel.app/best-gpu-for/qwen3.8/) and the [Qwen3.8-27B model page](https://whatcanirunai.vercel.app/models/qwen3.8-27b/).

## Other hardware for Qwen3.8-27B

* [Qwen3.8-27B on an RTX 3060 12GB](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3060-12gb/)Won’t fit
* [Qwen3.8-27B on an RTX 3070 Ti](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3070-ti/)Won’t fit
* [Qwen3.8-27B on an RTX 3090](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-3090/)Fits
* [Qwen3.8-27B on an RTX 4060 Ti 16GB](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-4060-ti-16gb/)Won’t fit
* [Qwen3.8-27B on an RTX 4070 Ti Super](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-4070-ti-super/)Won’t fit
* [Qwen3.8-27B on an RTX 4090](https://whatcanirunai.vercel.app/can-i-run/qwen3.8-27b/rtx-4090/)Fits

## Other models on an RTX 3080

* [Qwen3-32B on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/qwen3-32b/rtx-3080-10gb/)Won’t fit
* [Qwen3-30B-A3B on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/qwen3-30b-a3b/rtx-3080-10gb/)Won’t fit
* [Gemma 3 27B on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/gemma-3-27b-it/rtx-3080-10gb/)Won’t fit
* [Gemma 4 31B on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/gemma-4-31b-it/rtx-3080-10gb/)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
* [Mistral Small 24B on an RTX 3080](https://whatcanirunai.vercel.app/can-i-run/mistral-small-3.2-24b-instruct-2506/rtx-3080-10gb/)Won’t fit

## Quick answers

Can I run Qwen3.8-27B on an RTX 3080?

No. Qwen3.8-27B at Q4\_K\_M needs about 20.4 GB, and the GeForce RTX 3080 10GB has 10.0 GB usable.

How much VRAM does Qwen3.8-27B need?

About 20.4 GB at Q4\_K\_M, 34.9 GB at Q8\_0 and 64.9 GB at FP16, estimated for a \~4k context.

How fast is Qwen3.8-27B on an RTX 3080?

We do not estimate tokens/sec when the model does not fit. CPU offload is possible but slow.

What context length can I use?

The weights already do not fit, so longer context is not the limiting factor.

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