Can I Run GPT-OSS on NVIDIA GeForce RTX 3090?
作者： Jakub Rusinowski · 最后更新： 2026年8月15日
Yes, comfortably — GPT-OSS 20B at MXFP4 needs 12.1 GB at 8K context (11.1 GB weights + 1 GB KV cache/overhead), leaving ~11.9 GB of the NVIDIA GeForce RTX 3090's 24 GB, at ~174 tok/s (est.).
See what else this hardware can run →
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。
NVIDIA GeForce RTX 3090 Specs
| VRAM | 24 GB |
| Memory Bandwidth | 936 GB/s |
GPT-OSS 20B on the NVIDIA GeForce RTX 3090: VRAM by quantization
| Quant | VRAM needed | Fits 24 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 42.8 GB | ✗ No | — | — |
| Q8_0 | 23.2 GB | ✓ Yes | 32K | — |
| Q6_K | 18.2 GB | ✓ Yes | 128K | — |
| Q5_K_M | 15.8 GB | ✓ Yes | 128K | — |
| Q4_K_M | 13.6 GB | ✓ Yes | 128K | — |
| Q3_K_M | 9.9 GB | ✓ Yes | 128K | — |
| Q2_K | 7.9 GB | ✓ Yes | 128K | — |
Assumes an 8K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 24 GB. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
Context costs VRAM too. At MXFP4 on the NVIDIA GeForce RTX 3090: 8K 12.1 GB · 32K 12.7 GB · 128K 15.1 GB. Every window this model publishes still fits 24 GB.
No compatible GPU? GPT-OSS 20B on 16 GB of system RAM, CPU only: runs.
GPT-OSS Sizes That Fit the NVIDIA GeForce RTX 3090 (at 8K context)
| GPT-OSS 20B | MXFP4 · 12.1 GB at 8K context · ~174 tok/s (est.) |
At 2 hrs/day, buying (~$1,499) beats renting at $0.34/hr after about 6.1 years.
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
FAQ
Can I run GPT-OSS on the NVIDIA GeForce RTX 3090?
Yes, comfortably — GPT-OSS 20B at MXFP4 needs 12.1 GB at 8K context (11.1 GB weights + 1 GB KV cache/overhead), leaving ~11.9 GB of the NVIDIA GeForce RTX 3090's 24 GB, at ~174 tok/s (est.).
What's the best GPT-OSS size for the NVIDIA GeForce RTX 3090?
GPT-OSS 20B at MXFP4 quantization needs 12.1 GB at 8K context (11.1 GB weights + 1 GB KV cache/overhead), estimated ~174 tokens/sec.
Every GPT-OSS size on the NVIDIA GeForce RTX 3090
| Size | VRAM at 8K | Verdict | Speed |
|---|---|---|---|
| GPT-OSS 120B | 71.6 GB | Does not fit | — |
| GPT-OSS 20B | 12.1 GB | Runs | ~174 tok/s |
GPT-OSS on Other GPUs
- GPT-OSS on NVIDIA GeForce RTX 4090
- GPT-OSS on AMD Radeon RX 7900 XTX
- GPT-OSS on Apple M4 Pro
- GPT-OSS on Apple M3
- GPT-OSS on Apple M2
Popular Models on the NVIDIA GeForce RTX 3090
- DeepSeek R1 on NVIDIA GeForce RTX 3090
- Qwen 3 on NVIDIA GeForce RTX 3090
- Qwen 2.5 Family on NVIDIA GeForce RTX 3090
VRAM Tier
Troubleshooting
Buying Guide
← Can I Run It? | GPT-OSS Model Page | NVIDIA GeForce RTX 3090 GPU Page | VRAM calculator | Check Your Hardware