作者： Jakub Rusinowski · 最后更新： 2026年8月15日
Yes
Yes — GPT-OSS 20B at Q3_K_M needs about 10.1 GB of the 12 GB usable on RTX 3060 12 GB Desktop (12 GB VRAM, 32 GB RAM) (~1.9 GB spare), at ~115.3 tok/s (estimated), with room for about 32,768 tokens of context.
Confidence: high · Recommended quantization: Q3_K_M · Estimated speed: ~115.3 tok/s
or compare on Vast.ai from $0.35/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
| Usable memory for models | 12 GB |
| Memory bandwidth | 360 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 12 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 43 GB | ✗ No | — | — | 41.8 GB |
| Q8_0 | 23.4 GB | ✗ No | — | — | 22.2 GB |
| Q6_K | 18.3 GB | ✗ No | — | — | 17.1 GB |
| Q5_K_M | 16 GB | ✗ No | — | — | 14.8 GB |
| Q4_K_M | 13.8 GB | ✗ No | — | — | 12.6 GB |
| Q3_K_M | 10.1 GB | ✓ Yes | 32K | ~115.3 tok/s | 8.9 GB |
| Q2_K | 8.1 GB | ✓ Yes | 64K | ~135.6 tok/s | 6.9 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| GPT-oss 120B | 71.9 GB | ✗ Too large | — |
| GPT-OSS 20B | 13.8 GB | ✗ Too large | — |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes — GPT-OSS 20B at Q3_K_M needs about 10.1 GB of the 12 GB usable on RTX 3060 12 GB Desktop (12 GB VRAM, 32 GB RAM) (~1.9 GB spare), at ~115.3 tok/s (estimated), with room for about 32,768 tokens of context.
Q3_K_M — it needs about 10.1 GB of the 12 GB available, downloads as roughly 8.9 GB, and runs at an estimated 115.3 tokens/sec with up to 32K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving