# gpt-oss-20b vs Grok-2 (Dec 2024)

> Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 32.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-grok-2
- Last updated: 2026-10-11
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. gpt-oss-20b scores higher in 4 categories and Grok-2 (Dec 2024) in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-20b leads 39.4 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 11.5% for Grok-2 (Dec 2024).
- gpt-oss-20b has downloadable open weights; the other is API-only.

## Snapshot

| | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 32.5 | 33.7 |
| Rank | 255 | 239 |
| Context | 131K | — |
| Input $/M | $0.018 | — |
| Output $/M | $0.09 | — |
| Weights | Open | Proprietary |

## Coding

- gpt-oss-20b: 37.6 (#192)
- Grok-2 (Dec 2024): 33.3 (#258)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| WeirdML | 40.9% | 22.2% |
| LMArena Coding | 1306 | 1287 |
| SciCode | 34.4% | — |
| LiveBench Coding | — | 46.4% |
| ALE-Bench | 566.05 | — |

## Agentic & Tool Use

- gpt-oss-20b: 9.3 (#154)
- Grok-2 (Dec 2024): —

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| Terminal-Bench | 3.4% | — |

## Reasoning

- gpt-oss-20b: 19.3 (#261)
- Grok-2 (Dec 2024): 16.9 (#299)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1272 |
| DTBench | 68% | 65.2% |
| Epoch Capabilities Index | 137.82 | 130.48 |
| SimpleBench | — | 22.7% |
| Kagi LLM Benchmark | 53.2% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| LiveBench Reasoning | — | 54.8% |
| LiveBench Data Analysis | — | 54.5% |
| LMCA | 14.5% | — |
| LiveBench | — | 54.3% |

## Math

- gpt-oss-20b: 39.4 (#103)
- Grok-2 (Dec 2024): 20.8 (#284)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 11.5% |
| LMArena Math | 1317 | 1283 |
| Omni-MATH | 56.5% | — |
| LiveBench Math | — | 54.9% |
| MATH Level 5 | — | 63.5% |
| FrontierMath (Feb 2025 set) | — | 0.7% |

## Knowledge

- gpt-oss-20b: 34.6 (#195)
- Grok-2 (Dec 2024): 29.8 (#233)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| GPQA Diamond | 60.8% | 53.8% |
| LMArena Expert | 1258 | 1254 |
| MMLU-Pro | 74% | — |
| Confabulations | — | 20.1% |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- gpt-oss-20b: 42.2 (#197)
- Grok-2 (Dec 2024): 43.1 (#188)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Non-English | 1268 | 1282 |
| LMArena Chinese | 1314 | 1289 |
| LMArena German | 1255 | 1287 |
| LMArena Japanese | 1244 | 1244 |
| LMArena Korean | 1236 | 1237 |
| LMArena Russian | 1278 | 1286 |
| LMArena Spanish | 1267 | 1281 |
| LMArena French | — | 1318 |

## Instruction Following

- gpt-oss-20b: 61.8 (#240)
- Grok-2 (Dec 2024): 66.9 (#202)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Instruction Following | 1236 | 1270 |
| LiveBench Instruction Following | — | 69.6% |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#209)
- Grok-2 (Dec 2024): 38.8 (#190)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Longer Query | 1250 | 1276 |

## Writing & Preference

- gpt-oss-20b: 35.5 (#265)
- Grok-2 (Dec 2024): 48.6 (#198)

| Benchmark | gpt-oss-20b | Grok-2 (Dec 2024) |
|---|---|---|
| LMArena Text | 1287 | 1305 |
| LMArena Creative Writing | 1201 | 1284 |
| LMArena Multi-Turn | 1268 | 1290 |
| Short-Story Creative Writing | — | 63.6% |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
| LiveBench Language | — | 45.6% |

## FAQ

### Is gpt-oss-20b better than Grok-2 (Dec 2024)?

Grok-2 (Dec 2024) is the stronger model overall, scoring 33.7 to 32.5 on the Noometry Index.

### Is gpt-oss-20b or Grok-2 (Dec 2024) better for coding?

gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 33.3 in the Noometry coding category.

### How many benchmarks do gpt-oss-20b and Grok-2 (Dec 2024) share?

21 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Grok-2 (Dec 2024) has 34.