gpt-oss
gpt-oss-20b
OpenAI · open-permissive apache-2.0
| Parameters (B) | 21.512 |
|---|---|
| Architecture | moe |
| Layers / KV heads / head dim | 24 / 8 / 64 |
| Context window | 131072 |
| Modalities | in: text; out: text |
| GGUF | available |
| Commercial use | yes |
Minimum hardware (estimate)
At Q4_K_M and 8k context, total working set ≈ 14.0 GiB. On a machine with 64 GiB system RAM and no discrete GPU: CPU-only, slow. Limiting factor: No GPU VRAM; fits in system RAM but inference will be CPU-bound.
Scores
| Benchmark | Value | Uncertainty | Run by | Source |
|---|---|---|---|---|
| GPQA Diamond | 46.0% | unknown | third-party | epoch-ai |
| OTIS Mock AIME 2024-2025 | 53.9% | unknown | third-party | epoch-ai |
| Terminal-Bench | 3.4% | unknown | third-party | epoch-ai |
| Terminal-Bench | 3.4% | unknown | third-party | epoch-ai |
| Terminal-Bench | 3.1% | unknown | third-party | epoch-ai |
| Terminal-Bench | 3.1% | unknown | third-party | epoch-ai |
Provenance
- identity, license, architecture, parameter count: https://huggingface.co/openai/gpt-oss-20b (retrieved 2026-08-10)
HF: openai/gpt-oss-20b · snapshot 2026-08-10