Can gpt oss 20b run on the RX 9070?
Yes. At 8K context the best fit is Q8_0, using about 12.3 GB of the 15.2 GB usable on the RX 9070. Estimated speed ceiling: ~264 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 11.6 GB Fits | 11.7 GB Fits | 12.2 GB Fits | 14.5 GB Tight | 278 |
| 3-bit (Q3_K_M) • | 11.6 GB Fits | 11.7 GB Fits | 12.3 GB Fits | 14.5 GB Tight | 277 |
| 4-bit (Q4_K_M) • | 11.7 GB Fits | 11.8 GB Fits | 12.4 GB Fits | 14.6 GB Tight | 275 |
| 5-bit (Q5_K_M) • | 11.8 GB Fits | 11.9 GB Fits | 12.4 GB Fits | 14.7 GB Tight | 272 |
| 6-bit (Q6_K) • | 12.1 GB Fits | 12.2 GB Fits | 12.8 GB Fits | 15.0 GB Tight | 265 |
| 8-bit (Q8_0) • | 12.2 GB Fits | 12.3 GB Fits | 12.8 GB Fits | 15.1 GB Tight | 264 |
| 16-bit (F16/BF16) • | 13.7 GB Tight | 13.8 GB Tight | 14.4 GB Tight | 16.6 GB No | 231 |
• real GGUF file size from unsloth/gpt-oss-20b-GGUF. Speed is a bandwidth ceiling (640 GB/s), not a benchmark. Methodology.
More
gpt oss 20b on every GPU and Mac · Every model that runs on the RX 9070 · Other models that fit 16 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (openai/gpt-oss-20b), AMD specs. Data as of 2026-11-17.