# gpt-oss-20b vs Grok 4.6

> Grok 4.6 is the stronger model overall, scoring 56.8 to 32.1 on the Noometry Index. gpt-oss-20b costs 83× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-oss-20b-vs-grok-4-6
- Last updated: 2026-10-10
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. gpt-oss-20b scores higher in 0 categories and Grok 4.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.3 to 18.8.
- The biggest single-benchmark swing is Chess Puzzles: 4% for gpt-oss-20b and 40% for Grok 4.6.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $2 / $6 for Grok 4.6.
- Grok 4.6 accepts more context: 500K tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.

## Snapshot

| | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 32.1 | 56.8 |
| Rank | 249 | 19 |
| Context | 131K | 500K |
| Input $/M | $0.018 | $2 |
| Output $/M | $0.09 | $6 |
| Weights | Open | Proprietary |

## Coding

- gpt-oss-20b: 37.1 (#185)
- Grok 4.6: 58.0 (#14)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| SciCode | 34.4% | 56.5% |
| WeirdML | 40.9% | 67.3% |
| LMArena Coding | 1307 | 1464 |
| ALE-Bench | 566.05 | 1,508 |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| CursorBench | — | 41.4% |
| LMArena WebDev | — | 1620 |
| FrontierSWE | — | 25.3% |

## Agentic & Tool Use

- gpt-oss-20b: 9.2 (#146)
- Grok 4.6: 39.4 (#25)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| APEX-Agents | — | 65.3% |
| GDP.pdf | — | 17.2% |
| Vending-Bench 2 | — | 9,047 |

## Reasoning

- gpt-oss-20b: 18.8 (#251)
- Grok 4.6: 61.3 (#18)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| CritPt | 1.4% | 19.7% |
| Chess Puzzles | 4% | 40% |
| LMArena Hard Prompts | 1274 | 1445 |
| DTBench | 68% | 97.3% |
| LMCA | 14.5% | 48.5% |
| Epoch Capabilities Index | 137.81 | 156.48 |
| ARC-AGI-2 | — | 67.1% |
| SimpleBench | — | 75.9% |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 80% |
| ARC-AGI-1 | — | 87.5% |
| EBR-Bench | — | 30.5% |
| Mystery Game Puzzles | — | 34% |

## Math

- gpt-oss-20b: 37.6 (#93)
- Grok 4.6: 66.8 (#19)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 99.2% |
| LMArena Math | 1317 | 1416 |
| FrontierMath (Tiers 1-3) | — | 66% |
| FrontierMath Tier 4 | — | 31.7% |
| ProofBench | — | 51% |
| Omni-MATH | 56.5% | — |

## Knowledge

- gpt-oss-20b: 34.6 (#186)
- Grok 4.6: 63.3 (#18)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 60.8% | 94% |
| LMArena Expert | 1258 | 1463 |
| SimpleQA Verified | — | 49.3% |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |

## Multimodal

- gpt-oss-20b: —
- Grok 4.6: 43.9 (#18)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| LMArena Vision | — | 1265 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |

## Multilingual

- gpt-oss-20b: 42.4 (#192)
- Grok 4.6: 53.3 (#67)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1269 | 1422 |
| LMArena Chinese | 1311 | 1479 |
| LMArena German | 1249 | 1425 |
| LMArena Japanese | 1245 | 1383 |
| LMArena Korean | 1239 | 1395 |
| LMArena Russian | 1279 | 1427 |
| LMArena Spanish | 1267 | 1408 |
| LMArena French | — | 1458 |

## Instruction Following

- gpt-oss-20b: 62.0 (#235)
- Grok 4.6: 75.6 (#59)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1431 |
| IFEval | 73.2% | — |

## Long Context

- gpt-oss-20b: 37.9 (#203)
- Grok 4.6: 44.4 (#62)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1251 | 1453 |

## Writing & Preference

- gpt-oss-20b: 35.8 (#260)
- Grok 4.6: 62.5 (#77)

| Benchmark | gpt-oss-20b | Grok 4.6 |
|---|---|---|
| LMArena Text | 1288 | 1427 |
| LMArena Creative Writing | 1201 | 1430 |
| LMArena Multi-Turn | 1268 | 1422 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |

## FAQ

### Is gpt-oss-20b better than Grok 4.6?

Grok 4.6 is the stronger model overall, scoring 56.8 to 32.1 on the Noometry Index. gpt-oss-20b costs 83× less per token, which makes it the better buy when Grok 4.6's lead doesn't matter for your workload.

### Which is cheaper, gpt-oss-20b or Grok 4.6?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Grok 4.6 lists at $2 and $6.

### Is gpt-oss-20b or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.0 versus 37.1 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.6 does, with 500K tokens against 131K.

### How many benchmarks do gpt-oss-20b and Grok 4.6 share?

26 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Grok 4.6 has 49.