# gpt-oss-120b vs Kimi K2 (Jul 2025)

> Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 36.0 on the Noometry Index. gpt-oss-120b costs 14× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-120b-vs-kimi-k2
- Last updated: 2026-10-10
- Shared benchmarks: 36

## Summary

- They share 36 benchmarks with published results for both. gpt-oss-120b scores higher in 2 categories and Kimi K2 (Jul 2025) in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Kimi K2 (Jul 2025) leads 32.2 to 12.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 26% for gpt-oss-120b and 63.4% for Kimi K2 (Jul 2025).
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 131K.

## Snapshot

| | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 36.0 | 40.9 |
| Rank | 210 | 134 |
| Context | 131K | 262K |
| Input $/M | $0.037 | $0.57 |
| Output $/M | $0.17 | $2.30 |
| Weights | Open | Open |

## Coding

- gpt-oss-120b: 33.2 (#251)
- Kimi K2 (Jul 2025): 43.1 (#83)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 26% | 63.4% |
| Aider Polyglot | 41.8% | 59.1% |
| WeirdML | 48.2% | 42.8% |
| LMArena Coding | 1380 | 1399 |
| ALE-Bench | 575.62 | 597.5 |
| SciCode | 36% | — |
| GSO | — | 4.9% |
| AlgoTune | 1.41 | — |

## Agentic & Tool Use

- gpt-oss-120b: 12.1 (#145)
- Kimi K2 (Jul 2025): 32.2 (#59)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | 18.7% | 35.7% |
| METR Time Horizons | 56.6% | 59.2% |
| APEX-Agents | 4.4% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| Vending-Bench 2 | -21.53 | — |

## Reasoning

- gpt-oss-120b: 19.5 (#236)
- Kimi K2 (Jul 2025): 22.7 (#173)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 22.1% | 26.3% |
| Kagi LLM Benchmark | 58.6% | 64.4% |
| LMArena Hard Prompts | 1364 | 1384 |
| Epoch Capabilities Index | 139.94 | 146.02 |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| ForecastBench | — | 60.2 |

## Math

- gpt-oss-120b: 50.6 (#43)
- Kimi K2 (Jul 2025): 40.1 (#81)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 68.8% | 65.4% |
| LMArena Math | 1388 | 1396 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- gpt-oss-120b: 42.3 (#89)
- Kimi K2 (Jul 2025): 36.9 (#153)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 79.5% | 81.9% |
| Confabulations | 15.7% | 20.4% |
| Vectara Hallucination Rate | 14.2% | 17.9% |
| GPQA (HELM) | 68.4% | 65.3% |
| LMArena Expert | 1356 | 1363 |
| GPQA Diamond | 75.8% | — |

## Multilingual

- gpt-oss-120b: 48.2 (#142)
- Kimi K2 (Jul 2025): 49.8 (#125)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1350 | 1372 |
| LMArena Chinese | 1384 | 1416 |
| LMArena French | 1370 | 1379 |
| LMArena German | 1355 | 1386 |
| LMArena Japanese | 1332 | 1349 |
| LMArena Korean | 1283 | 1324 |
| LMArena Russian | 1343 | 1384 |
| LMArena Spanish | 1388 | 1383 |

## Instruction Following

- gpt-oss-120b: 69.5 (#168)
- Kimi K2 (Jul 2025): 71.3 (#151)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 83.6% | 85% |
| LMArena Instruction Following | 1318 | 1347 |

## Long Context

- gpt-oss-120b: 31.4 (#273)
- Kimi K2 (Jul 2025): 41.2 (#140)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | 44.4% | 66.7% |
| LMArena Longer Query | 1319 | 1353 |
| CL-bench | — | 17.6% |

## Writing & Preference

- gpt-oss-120b: 46.7 (#213)
- Kimi K2 (Jul 2025): 62.5 (#76)

| Benchmark | gpt-oss-120b | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1366 | 1380 |
| LMArena Creative Writing | 1274 | 1349 |
| Short-Story Creative Writing | 77.1% | 85.6% |
| EQ-Bench Creative Writing | 961 | 1666 |
| WildBench | 84.5% | 86.2% |
| LMArena Multi-Turn | 1340 | 1371 |

## FAQ

### Is gpt-oss-120b better than Kimi K2 (Jul 2025)?

Kimi K2 (Jul 2025) is the stronger model overall, scoring 40.9 to 36.0 on the Noometry Index. gpt-oss-120b costs 14× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-120b or Kimi K2 (Jul 2025)?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.

### Is gpt-oss-120b or Kimi K2 (Jul 2025) better for coding?

Kimi K2 (Jul 2025) scores higher on coding benchmarks: 43.1 versus 33.2 in the Noometry coding category.

### Which has the bigger context window?

Kimi K2 (Jul 2025) does, with 262K tokens against 131K.

### How many benchmarks do gpt-oss-120b and Kimi K2 (Jul 2025) share?

36 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Kimi K2 (Jul 2025) has 42.