# gpt-oss-120b vs Mistral Small

> gpt-oss-120b is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-oss-120b-vs-mistral-small
- Last updated: 2026-10-11
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. gpt-oss-120b scores higher in 5 categories and Mistral Small in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for gpt-oss-120b and 5.8% for Mistral Small.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.15 / $0.60 for Mistral Small.
- Mistral Small accepts more context: 262K tokens versus 131K.

## Snapshot

| | gpt-oss-120b | Mistral Small |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 36.3 | 33.4 |
| Rank | 217 | 243 |
| Context | 131K | 262K |
| Input $/M | $0.037 | $0.15 |
| Output $/M | $0.17 | $0.60 |
| Weights | Open | Open |

## Coding

- gpt-oss-120b: 33.5 (#256)
- Mistral Small: 34.0 (#247)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| SciCode | 36% | 26.5% |
| LMArena Coding | 1380 | 1362 |
| ALE-Bench | 575.62 | 497.62 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| WeirdML | 48.2% | — |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
| AlgoTune | 1.41 | — |

## Agentic & Tool Use

- gpt-oss-120b: 12.2 (#153)
- Mistral Small: 28.1 (#93)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| Berkeley Function Calling Leaderboard | — | 37.1% |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |

## Reasoning

- gpt-oss-120b: 20.0 (#245)
- Mistral Small: 19.8 (#250)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 58.6% | 37.8% |
| CritPt | 1.1% | 0% |
| LMArena Hard Prompts | 1364 | 1335 |
| DTBench | 76.3% | 70.9% |
| LMCA | 22.1% | 20.6% |
| SimpleBench | 22.1% | — |
| Chess Puzzles | 20% | — |
| LiveBench Reasoning | — | 44.8% |
| Mystery Game Puzzles | 2% | — |
| LiveBench Data Analysis | — | 53.7% |
| Surface Evolver Bench | 25% | — |
| Epoch Capabilities Index | 139.93 | — |
| LiveBench | — | 44% |

## Math

- gpt-oss-120b: 52.5 (#50)
- Mistral Small: 16.4 (#293)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 5.8% |
| LMArena Math | 1389 | 1341 |
| Omni-MATH | 68.8% | — |
| LiveBench Math | — | 39.9% |
| MATH Level 5 | — | 46.8% |

## Knowledge

- gpt-oss-120b: 42.4 (#96)
- Mistral Small: 31.0 (#222)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| GPQA Diamond | 75.8% | 47.5% |
| Vectara Hallucination Rate | 14.2% | 5.1% |
| LMArena Expert | 1356 | 1291 |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| GPQA (HELM) | 68.4% | — |
| MMLU | — | 68.7% |

## Multimodal

- gpt-oss-120b: —
- Mistral Small: 33.5 (#96)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |

## Multilingual

- gpt-oss-120b: 48.0 (#147)
- Mistral Small: 45.5 (#169)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| LMArena Non-English | 1351 | 1315 |
| LMArena Chinese | 1385 | 1340 |
| LMArena French | 1369 | 1337 |
| LMArena German | 1353 | 1340 |
| LMArena Japanese | 1331 | 1275 |
| LMArena Korean | 1282 | 1259 |
| LMArena Russian | 1343 | 1324 |
| LMArena Spanish | 1389 | 1346 |

## Instruction Following

- gpt-oss-120b: 69.3 (#173)
- Mistral Small: 66.4 (#209)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1318 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 83.6% | — |

## Long Context

- gpt-oss-120b: 31.4 (#278)
- Mistral Small: 40.4 (#156)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1319 | 1327 |
| Fiction.LiveBench | 44.4% | — |

## Writing & Preference

- gpt-oss-120b: 46.5 (#217)
- Mistral Small: 52.5 (#171)

| Benchmark | gpt-oss-120b | Mistral Small |
|---|---|---|
| LMArena Text | 1365 | 1338 |
| LMArena Creative Writing | 1275 | 1305 |
| LMArena Multi-Turn | 1340 | 1344 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| LiveBench Language | — | 30.5% |

## FAQ

### Is gpt-oss-120b better than Mistral Small?

gpt-oss-120b is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index.

### Which is cheaper, gpt-oss-120b or Mistral Small?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Mistral Small lists at $0.15 and $0.60.

### Is gpt-oss-120b or Mistral Small better for coding?

They score almost the same on coding (33.5 vs 34.0); test both on your own repository before choosing.

### Which has the bigger context window?

Mistral Small does, with 262K tokens against 131K.

### How many benchmarks do gpt-oss-120b and Mistral Small share?

26 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Mistral Small has 39.