Can I Run GPT-OSS on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
作者： Jakub Rusinowski · 最后更新： 2026年8月15日
Yes, but it is tight
Yes, but it is tight — GPT-OSS 20B at Q8_0 needs about 23.2 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~0.8 GB before the runtime starts swapping. Expect ~134 tok/s (estimated), with room for about 32,768 tokens of context.
Confidence: medium · Recommended quantization: Q8_0 · Estimated speed: ~134 tok/s
Get a personalized upgrade path →
或在 Vast.ai 比较，低至 $0.35/小时 (typical low · varies)
作为亚马逊联盟成员，我们从符合条件的购买中获得收入。云 GPU 链接为推荐链接——我们可能获得佣金，您无需额外付费。
RTX 4090 Desktop (24 GB VRAM, 64 GB RAM) — what it gives a model
| Usable memory for models | 24 GB |
| Memory bandwidth | 1008 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
GPT-OSS on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM): memory by quantization
| Quant | Memory needed | Fits 24 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 42.8 GB | ✗ No | — | — | 41.8 GB |
| Q8_0 | 23.2 GB | ✓ Yes | 32K | ~134 tok/s | 22.2 GB |
| Q6_K | 18.2 GB | ✓ Yes | 128K | ~158.5 tok/s | 17.1 GB |
| Q5_K_M | 15.8 GB | ✓ Yes | 128K | ~173 tok/s | 14.8 GB |
| Q4_K_M | 13.6 GB | ✓ Yes | 128K | ~189.4 tok/s | 12.6 GB |
| Q3_K_M | 9.9 GB | ✓ Yes | 128K | ~225.5 tok/s | 8.9 GB |
| Q2_K | 7.9 GB | ✓ Yes | 128K | ~251.9 tok/s | 6.9 GB |
Which GPT-OSS sizes fit
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| GPT-OSS 120B | 71.6 GB | ✗ Too large | — |
| GPT-OSS 20B | 13.6 GB | ✓ Fits | ~189.4 tok/s |
What to watch out for
- Only ~0.8 GB of headroom at Q8_0: a longer context or a second application can push this into swapping.
- 1 larger variant of GPT-OSS does not fit and would need CPU offload or different hardware.
RTX 4090 desktop limitations
- 24 GB is the sweet spot for 27–32B models at Q4; 70B needs offload or a second card.
Recommended setup
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 24 GB of VRAM on the NVIDIA GeForce RTX 4090 at 1008 GB/s.
- 64 GB of system RAM available for CPU offload when a model exceeds VRAM.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is computed from this model's published attention configuration.
FAQ
Can I run GPT-OSS on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Yes, but it is tight — GPT-OSS 20B at Q8_0 needs about 23.2 GB of the 24 GB usable on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM), leaving only ~0.8 GB before the runtime starts swapping. Expect ~134 tok/s (estimated), with room for about 32,768 tokens of context.
Which quantization of GPT-OSS should I use on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Q8_0 — it needs about 23.2 GB of the 24 GB available, downloads as roughly 22.2 GB, and runs at an estimated 134 tokens/sec with up to 32K of context.
What limits GPT-OSS on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)?
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Which runtime should I use?
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Other Computers
Other Models on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- Granite 3.0 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- IBM Granite 4.0 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- IBM Granite 4.1 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- IBM Granite 4.2 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
- InternLM 3 on RTX 4090 Desktop (24 GB VRAM, 64 GB RAM)
GPT-OSS on GPUs
- GPT-OSS on NVIDIA GeForce RTX 5080
- GPT-OSS on NVIDIA GeForce RTX 5070 Ti
- GPT-OSS on NVIDIA GeForce RTX 5070
- GPT-OSS on NVIDIA GeForce RTX 5060 Ti 16GB