GPT OSS 120B
| Released | 2025-08-05 |
|---|---|
| Price per 1M | $0.15 in · $0.60 out |
| Overall score | 44.3 · rank 162 |
| Context window | 131,072 tokens |
| Provider | OpenAI |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 44.3 | 162 | 20 |
| Overall: Coding | 43.7 | 166 | 13 |
| Overall: Reasoning | 47.3 | 139 | 7 |
| Overall: Agentic work | 43.0 | 171 | 13 |
| Overall: Frontend | 44.1 | 146 | 4 |
| Coding | 44.1 | 164 | 4 |
| Agents | 42.1 | 150 | 4 |
| Reasoning | 44.7 | 156 | 5 |
| Math | 53.5 | 83 | 1 |
| Knowledge | 42.2 | 133 | 1 |
| General | 41.9 | 155 | 5 |
| Value vs price | +0.5 | 116 |
Configurations
| Configuration | Balanced | Benchmarks |
|---|---|---|
| Unspecified setting | 44.6 ±2.3 | 15 |
| High | 42.3 ±5.8 | 6 |
| Low | 41.8 ±5.2 | 2 |
| Medium | 31.9 ±12.0 | 1 |
Evidence
Coding 44.1 ±5.2
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Aider Polyglot | 41.8% | Epoch AI | 2026-11-11 |
| SWE-bench Verified (Vals run) | 33.6% | Vals AI | 2026-09-01 |
| SciCode | 36.0% | Epoch AI | 2026-11-11 |
| LiveCodeBench (Vals run) | 83.2% | Vals AI | 2026-09-01 |
Agents 42.1 ±3.8
| Benchmark | Result | Source | Date |
|---|---|---|---|
| Terminal-Bench 2.0 | 18.7% | Epoch AI | 2025-11-01 |
| APEX-Agents | 4.4% | Epoch AI | 2026-11-11 |
| METR task suite | 56.6% | Epoch AI | 2026-11-11 |
| Vending-Bench 2 | $-22 | Epoch AI | 2026-11-11 |
Reasoning 44.7 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| CritPt | 1.1% | Epoch AI | 2026-11-11 |
| GPQA Diamond | 75.8% | Epoch AI | 2025-12-11 |
| GPQA Diamond (Vals run) | 78.5% | Vals AI | 2026-09-01 |
| Kagi Reasoning Benchmark | 58.6% | Kagi | 2026-11-11 |
| SimpleBench | 22.1% | SimpleBench | 2026-11-11 |
| SimpleBench | 22.1% | Epoch AI | 2026-11-11 |
| Mystery Game Puzzles | 2.0% | Epoch AI | 2026-08-27 |
Math 53.5 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | Epoch AI | 2025-12-11 |
Knowledge 42.2 ±4.7
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MMLU-Pro (Vals run) | 79.2% | Vals AI | 2026-09-01 |
General 41.9 ±4.1
| Benchmark | Result | Source | Date |
|---|---|---|---|
| LLM Debate Benchmark | 1,263 | Lech Mazur | 2026-10-26 |
| EQ-Bench Creative Writing v3 | 956 | EQ-Bench | 2026-11-11 |
| LMArena Text | 1,352 | LMArena | 2026-11-09 |
| Creative Story-Writing (Lech Mazur) | 11.0% | Lech Mazur | 2026-10-23 |
| EQ-Bench Longform Creative Writing | 35.8% | EQ-Bench | 2026-11-11 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| CritPt | 0.0% | Epoch AI | A stronger configuration result was used |
| Mystery Game Puzzles | 0.0% | Epoch AI | A stronger configuration result was used |
| SciCode | 34.0% | Epoch AI | A stronger configuration result was used |
| Terminal-Bench 2.0 | 14.2% | Epoch AI | A stronger configuration result was used |
Names matched to this model
gpt-oss-120b