作者： Jakub Rusinowski · 最后更新： 2026年8月15日
Yes, but it's tight — GPT-OSS 20B at MXFP4 needs 10.625 GB of the NVIDIA GeForce RTX 3060 (12GB)'s 12 GB, leaving only ~1.4 GB, ~20 tok/s (est.).
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.
| VRAM | 12 GB |
| Memory Bandwidth | 360 GB/s |
| Quant | VRAM needed | Fits 12 GB? | Max context |
|---|---|---|---|
| F16 | 41 GB | ✗ No | — |
| Q8_0 | 22.3 GB | ✗ No | — |
| Q6_K | 17.4 GB | ✗ No | — |
| Q5_K_M | 15.2 GB | ✗ No | — |
| Q4_K_M | 13.1 GB | ✗ No | — |
| Q3_K_M | 9.5 GB | ✓ Yes | 32K |
| Q2_K | 7.6 GB | ✓ Yes | 64K |
VRAM needed assumes a 4K-token context with an f16 KV cache; “Max context” is the largest window that still fits in 12 GB. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
| GPT-OSS 20B | MXFP4 · 10.625 GB · ~20 tok/s (est.) |
At 2 hrs/day, buying (~$329) beats renting at $0.34/hr after about 16 months.
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.
Yes, but it's tight — GPT-OSS 20B at MXFP4 needs 10.625 GB of the NVIDIA GeForce RTX 3060 (12GB)'s 12 GB, leaving only ~1.4 GB, ~20 tok/s (est.).
GPT-OSS 20B at MXFP4 quantization (10.625 GB), estimated ~20 tokens/sec.
| Size | VRAM | Verdict | Speed |
|---|---|---|---|
| GPT-oss 120B | 72.5 GB | Does not fit | — |
| GPT-OSS 20B | 10.6 GB | Runs | ~20 tok/s |
← Can I Run It? | GPT-OSS Model Page | NVIDIA GeForce RTX 3060 (12GB) GPU Page | Check Your Hardware