# gpt-oss-120b vs Grok 4.5

> Grok 4.5 is the stronger model overall, scoring 55.0 to 36.3 on the Noometry Index. gpt-oss-120b costs 43× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-120b-vs-grok-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. gpt-oss-120b scores higher in 0 categories and Grok 4.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 20.0.
- The biggest single-benchmark swing is APEX-Agents: 4.4% for gpt-oss-120b and 56.2% for Grok 4.5.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $2 / $6 for Grok 4.5.
- Grok 4.5 accepts more context: 500K tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.

## Snapshot

| | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 36.3 | 55.0 |
| Rank | 217 | 25 |
| Context | 131K | 500K |
| Input $/M | $0.037 | $2 |
| Output $/M | $0.17 | $6 |
| Weights | Open | Proprietary |

## Coding

- gpt-oss-120b: 33.5 (#256)
- Grok 4.5: 52.2 (#35)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| SciCode | 36% | 54.1% |
| WeirdML | 48.2% | 46.4% |
| LMArena Coding | 1380 | 1474 |
| ALE-Bench | 575.62 | 1,309 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1553 |
| AlgoTune | 1.41 | — |

## Agentic & Tool Use

- gpt-oss-120b: 12.2 (#153)
- Grok 4.5: 44.4 (#17)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| APEX-Agents | 4.4% | 56.2% |
| Vending-Bench 2 | -21.53 | 3,887 |
| Terminal-Bench | 18.7% | — |
| τ²-bench Banking | — | 47.9% |
| PostTrainBench | — | 23.4% |
| GBAEval | — | 65.4% |
| GDP.pdf | — | 14% |
| LMArena Search | — | 1213 |
| METR Time Horizons | 56.6% | — |

## Reasoning

- gpt-oss-120b: 20.0 (#245)
- Grok 4.5: 56.1 (#25)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| SimpleBench | 22.1% | 70% |
| Kagi LLM Benchmark | 58.6% | 83.5% |
| CritPt | 1.1% | 15.4% |
| Chess Puzzles | 20% | 36% |
| LMArena Hard Prompts | 1364 | 1462 |
| DTBench | 76.3% | 96.5% |
| LMCA | 22.1% | 45.2% |
| Surface Evolver Bench | 25% | 74.4% |
| Epoch Capabilities Index | 139.93 | 153.92 |
| ARC-AGI-2 | — | 52.6% |
| NYT Connections (extended) | — | 79.9% |
| ARC-AGI-1 | — | 87.2% |
| Mystery Game Puzzles | 2% | — |

## Math

- gpt-oss-120b: 52.5 (#50)
- Grok 4.5: 60.9 (#35)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 97.8% |
| LMArena Math | 1389 | 1459 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 24.4% |
| ProofBench | — | 31% |
| Omni-MATH | 68.8% | — |

## Knowledge

- gpt-oss-120b: 42.4 (#96)
- Grok 4.5: 62.3 (#24)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| GPQA Diamond | 75.8% | 93.4% |
| LMArena Expert | 1356 | 1466 |
| SimpleQA Verified | — | 48.3% |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |

## Multimodal

- gpt-oss-120b: —
- Grok 4.5: 37.6 (#72)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| LMArena Vision | — | 1288 |
| Blueprint-Bench 2 | — | 27.3% |
| Furniture Assembly | — | 22.5% |
| LMArena Document | — | 1452 |

## Multilingual

- gpt-oss-120b: 48.0 (#147)
- Grok 4.5: 54.4 (#42)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1351 | 1440 |
| LMArena Chinese | 1385 | 1496 |
| LMArena French | 1369 | 1456 |
| LMArena German | 1353 | 1446 |
| LMArena Japanese | 1331 | 1428 |
| LMArena Korean | 1282 | 1404 |
| LMArena Russian | 1343 | 1448 |
| LMArena Spanish | 1389 | 1450 |

## Instruction Following

- gpt-oss-120b: 69.3 (#173)
- Grok 4.5: 76.0 (#48)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1318 | 1446 |
| IFEval | 83.6% | — |

## Long Context

- gpt-oss-120b: 31.4 (#278)
- Grok 4.5: 44.8 (#56)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1319 | 1463 |
| Fiction.LiveBench | 44.4% | — |

## Writing & Preference

- gpt-oss-120b: 46.5 (#217)
- Grok 4.5: 65.8 (#42)

| Benchmark | gpt-oss-120b | Grok 4.5 |
|---|---|---|
| LMArena Text | 1365 | 1448 |
| LMArena Creative Writing | 1275 | 1442 |
| EQ-Bench Creative Writing | 961 | 1579 |
| LMArena Multi-Turn | 1340 | 1456 |
| Short-Story Creative Writing | 77.1% | — |
| WildBench | 84.5% | — |

## FAQ

### Is gpt-oss-120b better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 36.3 on the Noometry Index. gpt-oss-120b costs 43× less per token, which makes it the better buy when Grok 4.5's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-120b or Grok 4.5?

gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Grok 4.5 lists at $2 and $6.

### Is gpt-oss-120b or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 33.5 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.5 does, with 500K tokens against 131K.

### How many benchmarks do gpt-oss-120b and Grok 4.5 share?

33 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Grok 4.5 has 52.