# gpt-oss-120b vs o3-pro

> o3-pro is the stronger model overall, scoring 42.9 to 36.3 on the Noometry Index. gpt-oss-120b costs 498× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-120b-vs-o3-pro
- Last updated: 2026-10-10
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. gpt-oss-120b scores higher in 1 category and o3-pro in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 31.4.
- The biggest single-benchmark swing is Fiction.LiveBench: 44.4% for gpt-oss-120b and 97.2% for o3-pro.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $20 / $80 for o3-pro.
- o3-pro accepts more context: 200K tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.

## Snapshot

| | gpt-oss-120b | o3-pro |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 36.3 | 42.9 |
| Rank | 217 | 105 |
| Context | 131K | 200K |
| Input $/M | $0.037 | $20 |
| Output $/M | $0.17 | $80 |
| Weights | Open | Proprietary |

## Coding

- gpt-oss-120b: 33.5 (#256)
- o3-pro: 55.5 (#24)

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Aider Polyglot | 41.8% | 84.9% |
| WeirdML | 48.2% | 58.2% |
| SWE-bench Verified (bash only) | 26% | — |
| SciCode | 36% | — |
| LMArena Coding | 1380 | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |

## Agentic & Tool Use

- gpt-oss-120b: 12.2 (#153)
- o3-pro: —

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |

## Reasoning

- gpt-oss-120b: 20.0 (#245)
- o3-pro: 23.8 (#171)

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Kagi LLM Benchmark | 58.6% | 72.1% |
| DTBench | 76.3% | 86.9% |
| LMCA | 22.1% | 38.5% |
| Epoch Capabilities Index | 139.93 | 147.42 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | 22.1% | — |
| ARC-AGI-1 | — | 59.3% |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| LMArena Hard Prompts | 1364 | — |
| Mystery Game Puzzles | 2% | — |
| Surface Evolver Bench | 25% | — |

## Math

- gpt-oss-120b: 52.5 (#50)
- o3-pro: —

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
| LMArena Math | 1389 | — |

## Knowledge

- gpt-oss-120b: 42.4 (#96)
- o3-pro: 29.5 (#238)

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Confabulations | 15.7% | 14.2% |
| Vectara Hallucination Rate | 14.2% | 23.3% |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| GPQA (HELM) | 68.4% | — |
| LMArena Expert | 1356 | — |

## Multilingual

- gpt-oss-120b: 48.0 (#147)
- o3-pro: —

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| LMArena Non-English | 1351 | — |
| LMArena Chinese | 1385 | — |
| LMArena French | 1369 | — |
| LMArena German | 1353 | — |
| LMArena Japanese | 1331 | — |
| LMArena Korean | 1282 | — |
| LMArena Russian | 1343 | — |
| LMArena Spanish | 1389 | — |

## Instruction Following

- gpt-oss-120b: 69.3 (#173)
- o3-pro: —

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| IFEval | 83.6% | — |
| LMArena Instruction Following | 1318 | — |

## Long Context

- gpt-oss-120b: 31.4 (#278)
- o3-pro: 72.2 (#1)

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Fiction.LiveBench | 44.4% | 97.2% |
| LMArena Longer Query | 1319 | — |

## Writing & Preference

- gpt-oss-120b: 46.5 (#217)
- o3-pro: 57.1 (#133)

| Benchmark | gpt-oss-120b | o3-pro |
|---|---|---|
| Short-Story Creative Writing | 77.1% | 84.4% |
| LMArena Text | 1365 | — |
| LMArena Creative Writing | 1275 | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| LMArena Multi-Turn | 1340 | — |

## FAQ

### Is gpt-oss-120b better than o3-pro?

o3-pro is the stronger model overall, scoring 42.9 to 36.3 on the Noometry Index. gpt-oss-120b costs 498× less per token, which makes it the better buy when o3-pro's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-120b or o3-pro?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; o3-pro lists at $20 and $80.

### Is gpt-oss-120b or o3-pro better for coding?

o3-pro scores higher on coding benchmarks: 55.5 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

o3-pro does, with 200K tokens against 131K.

### How many benchmarks do gpt-oss-120b and o3-pro share?

10 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and o3-pro has 12.