# Grok 4.6 vs o3

> Grok 4.6 is the stronger model overall, scoring 56.9 to 47.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-6-vs-o3
- Last updated: 2026-10-10
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. Grok 4.6 scores higher in 8 categories and o3 in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.6 leads 61.4 to 32.0.
- The biggest single-benchmark swing is ARC-AGI-2: 67.1% for Grok 4.6 and 6.5% for o3.
- Grok 4.6 is cheaper at $2 / $6 per million input/output tokens, against $2 / $8 for o3.
- Grok 4.6 accepts more context: 500K tokens versus 200K.

## Snapshot

| | Grok 4.6 | o3 |
|---|---|---|
| Provider | xAI | OpenAI |
| Noometry Index | 56.9 | 47.5 |
| Rank | 21 | 61 |
| Context | 500K | 200K |
| Input $/M | $2 | $2 |
| Output $/M | $6 | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Grok 4.6: 58.5 (#16)
- o3: 46.8 (#64)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| WeirdML | 67.3% | 52.4% |
| LMArena Coding | 1465 | 1408 |
| ALE-Bench | 1,508 | 933.55 |
| SWE-bench Verified | — | 62.3% |
| DeepSWE | 67.5% | — |
| FrontierCode | 48% | — |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| CursorBench | 41.4% | — |
| LMArena WebDev | 1617 | — |
| FrontierSWE | 25.3% | — |
| SciCode | 56.5% | — |
| GSO | — | 8.8% |
| CadEval | — | 74% |

## Agentic & Tool Use

- Grok 4.6: 39.4 (#27)
- o3: 34.5 (#44)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| APEX-Agents | 65.3% | — |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| GDP.pdf | 17.2% | — |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
| Vending-Bench 2 | 9,047 | — |

## Reasoning

- Grok 4.6: 61.4 (#20)
- o3: 32.0 (#78)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| ARC-AGI-2 | 67.1% | 6.5% |
| SimpleBench | 75.9% | 53.1% |
| ARC-AGI-1 | 87.5% | 60.8% |
| CritPt | 19.7% | 1.4% |
| Chess Puzzles | 40% | 38% |
| LMArena Hard Prompts | 1447 | 1402 |
| Mystery Game Puzzles | 34% | 29% |
| DTBench | 97.3% | 84.8% |
| LMCA | 48.5% | 39.7% |
| Epoch Capabilities Index | 156.44 | 146.86 |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 80% | — |
| EnigmaEval | — | 13.1% |
| EBR-Bench | 30.5% | — |
| ForecastBench | — | 62.5 |

## Math

- Grok 4.6: 67.0 (#24)
- o3: 50.2 (#58)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 66% | 33.3% |
| OTIS Mock AIME 2024-2025 | 99.2% | 84.4% |
| LMArena Math | 1423 | 1426 |
| FrontierMath Tier 4 | 31.7% | — |
| ProofBench | 51% | — |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 18.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Grok 4.6: 63.3 (#20)
- o3: 54.6 (#52)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| GPQA Diamond | 94% | 81.8% |
| SimpleQA Verified | 49.3% | 49.4% |
| LMArena Expert | 1467 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| GPQA (HELM) | — | 75.3% |

## Multimodal

- Grok 4.6: 43.6 (#23)
- o3: 41.4 (#36)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| LMArena Vision | 1263 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
| Blueprint-Bench 2 | 33.2% | — |
| Furniture Assembly | 40% | — |
| LMArena Document | 1452 | — |

## Multilingual

- Grok 4.6: 53.0 (#74)
- o3: 51.7 (#105)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| LMArena Non-English | 1420 | 1401 |
| LMArena Chinese | 1480 | 1437 |
| LMArena French | 1461 | 1430 |
| LMArena German | 1431 | 1420 |
| LMArena Japanese | 1376 | 1403 |
| LMArena Korean | 1397 | 1370 |
| LMArena Russian | 1422 | 1406 |
| LMArena Spanish | 1404 | 1395 |

## Instruction Following

- Grok 4.6: 75.4 (#63)
- o3: 72.8 (#127)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| LMArena Instruction Following | 1431 | 1368 |
| IFEval | — | 86.9% |

## Long Context

- Grok 4.6: 44.5 (#66)
- o3: 53.3 (#6)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| LMArena Longer Query | 1454 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |

## Writing & Preference

- Grok 4.6: 62.3 (#80)
- o3: 63.5 (#64)

| Benchmark | Grok 4.6 | o3 |
|---|---|---|
| LMArena Text | 1428 | 1410 |
| LMArena Creative Writing | 1428 | 1359 |
| LMArena Multi-Turn | 1425 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |

## FAQ

### Is Grok 4.6 better than o3?

Grok 4.6 is the stronger model overall, scoring 56.9 to 47.5 on the Noometry Index.

### Which is cheaper, Grok 4.6 or o3?

Grok 4.6 is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; o3 lists at $2 and $8.

### Is Grok 4.6 or o3 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 46.8 in the Noometry coding category.

### Which has the bigger context window?

Grok 4.6 does, with 500K tokens against 200K.

### How many benchmarks do Grok 4.6 and o3 share?

33 benchmarks have published results for both models. Grok 4.6 has 49 scored results on Noometry and o3 has 63.