# GPT-5.5 vs gpt-oss-120b

> GPT-5.5 is the stronger model overall, scoring 63.4 to 36.3 on the Noometry Index. gpt-oss-120b costs 160× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-5-vs-gpt-oss-120b
- Last updated: 2026-10-12
- Shared benchmarks: 36

## Summary

- They share 36 benchmarks with published results for both. GPT-5.5 scores higher in 9 categories and gpt-oss-120b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 20.0.
- The biggest single-benchmark swing is Terminal-Bench: 84.7% for GPT-5.5 and 18.7% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 63.4 | 36.3 |
| Rank | 9 | 217 |
| Context | 1.05M | 131K |
| Input $/M | $5 | $0.037 |
| Output $/M | $30 | $0.17 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.5: 58.2 (#17)
- gpt-oss-120b: 33.5 (#256)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| SciCode | 56.1% | 36% |
| WeirdML | 84.9% | 48.2% |
| LMArena Coding | 1494 | 1380 |
| ALE-Bench | 1,943 | 575.62 |
| SWE-bench Verified | 80.6% | — |
| DeepSWE | 67% | — |
| FrontierCode | 43% | — |
| SWE-bench Verified (bash only) | — | 26% |
| Aider Polyglot | — | 41.8% |
| LMArena WebDev | 1513 | — |
| GSO | 40.2% | — |
| MirrorCode | 10% | — |
| AlgoTune | — | 1.41 |

## Agentic & Tool Use

- GPT-5.5: 50.7 (#6)
- gpt-oss-120b: 12.2 (#153)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | 84.7% | 18.7% |
| APEX-Agents | 55.1% | 4.4% |
| Vending-Bench 2 | 7,524 | -21.53 |
| OSWorld 2.0 | 13% | — |
| Remote Labor Index | 6.3% | — |
| τ²-bench Banking | 44.6% | — |
| DeepResearch Bench | 54% | — |
| PostTrainBench | 27.2% | — |
| ExploitBench | 47.4% | — |
| GBAEval | 53.2% | — |
| GDP.pdf | 26% | — |
| LMArena Search | 1242 | — |
| METR Time Horizons | — | 56.6% |

## Reasoning

- GPT-5.5: 72.8 (#11)
- gpt-oss-120b: 20.0 (#245)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| SimpleBench | 69% | 22.1% |
| Kagi LLM Benchmark | 88.8% | 58.6% |
| CritPt | 27.1% | 1.1% |
| Chess Puzzles | 54% | 20% |
| LMArena Hard Prompts | 1489 | 1364 |
| Mystery Game Puzzles | 56% | 2% |
| DTBench | 96% | 76.3% |
| LMCA | 54.3% | 22.1% |
| Surface Evolver Bench | 88.1% | 25% |
| Epoch Capabilities Index | 159.1 | 139.93 |
| ARC-AGI-2 | 85% | — |
| NYT Connections (extended) | 96.2% | — |
| ARC-AGI-1 | 95% | — |
| EBR-Bench | 34.3% | — |
| Bench to the Future 3 | 0.14 | — |
| ForecastBench | 60.6 | — |

## Math

- GPT-5.5: 81.7 (#11)
- gpt-oss-120b: 52.5 (#50)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 88.9% |
| LMArena Math | 1486 | 1389 |
| FrontierMath (Tiers 1-3) | 85.3% | — |
| FrontierMath Tier 4 | 72.5% | — |
| MathArena Final-Answer Competitions | 94.3% | — |
| ProofBench | 50% | — |
| Omni-MATH | — | 68.8% |
| FrontierMath (Feb 2025 set) | 51.7% | — |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | 35.4% | — |

## Knowledge

- GPT-5.5: 64.4 (#17)
- gpt-oss-120b: 42.4 (#96)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 94% | 75.8% |
| Vectara Hallucination Rate | 9.3% | 14.2% |
| LMArena Expert | 1508 | 1356 |
| SimpleQA Verified | 63% | — |
| MMLU-Pro | — | 79.5% |
| Confabulations | — | 15.7% |
| GPQA (HELM) | — | 68.4% |

## Multimodal

- GPT-5.5: 46.9 (#12)
- gpt-oss-120b: —

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1297 | — |
| Blueprint-Bench 2 | 36.2% | — |
| Furniture Assembly | 44.2% | — |
| LMArena Document | 1486 | — |

## Multilingual

- GPT-5.5: 56.4 (#20)
- gpt-oss-120b: 48.0 (#147)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1467 | 1351 |
| LMArena Chinese | 1533 | 1385 |
| LMArena French | 1486 | 1369 |
| LMArena German | 1480 | 1353 |
| LMArena Japanese | 1498 | 1331 |
| LMArena Korean | 1460 | 1282 |
| LMArena Russian | 1473 | 1343 |
| LMArena Spanish | 1468 | 1389 |

## Instruction Following

- GPT-5.5: 77.5 (#18)
- gpt-oss-120b: 69.3 (#173)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| LMArena Instruction Following | 1479 | 1318 |
| IFEval | — | 83.6% |

## Long Context

- GPT-5.5: 48.3 (#12)
- gpt-oss-120b: 31.4 (#278)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1484 | 1319 |
| Fiction.LiveBench | — | 44.4% |
| CL-bench Life | 22.2% | — |

## Writing & Preference

- GPT-5.5: 72.7 (#13)
- gpt-oss-120b: 46.5 (#217)

| Benchmark | GPT-5.5 | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1472 | 1365 |
| LMArena Creative Writing | 1455 | 1275 |
| EQ-Bench Creative Writing | 1844 | 961 |
| LMArena Multi-Turn | 1476 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| WildBench | — | 84.5% |
| EQ-Bench 4 | 1315 | — |

## FAQ

### Is GPT-5.5 better than gpt-oss-120b?

GPT-5.5 is the stronger model overall, scoring 63.4 to 36.3 on the Noometry Index. gpt-oss-120b costs 160× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.5 or gpt-oss-120b?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; GPT-5.5 lists at $5 and $30.

### Is GPT-5.5 or gpt-oss-120b better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.2 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 131K.

### How many benchmarks do GPT-5.5 and gpt-oss-120b share?

36 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and gpt-oss-120b has 48.