# Claude 3 Haiku vs Grok 4.6

> Grok 4.6 is the stronger model overall, scoring 56.9 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-grok-4-6
- Last updated: 2026-10-12
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Grok 4.6 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4.6 leads 67.0 to 9.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 99.2% for Grok 4.6.

## Snapshot

| | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 25.9 | 56.9 |
| Rank | 340 | 21 |
| Context | — | 500K |
| Input $/M | — | $2 |
| Output $/M | — | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- Grok 4.6: 58.5 (#16)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| WeirdML | 9.8% | 67.3% |
| LMArena Coding | 1199 | 1465 |
| DeepSWE | — | 67.5% |
| FrontierCode | — | 48% |
| CursorBench | — | 41.4% |
| LMArena WebDev | — | 1617 |
| FrontierSWE | — | 25.3% |
| SciCode | — | 56.5% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 1,508 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Grok 4.6: 39.4 (#27)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| APEX-Agents | — | 65.3% |
| GDP.pdf | — | 17.2% |
| Vending-Bench 2 | — | 9,047 |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- Grok 4.6: 61.4 (#20)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1447 |
| DTBench | 50.1% | 97.3% |
| LMCA | 8.8% | 48.5% |
| Epoch Capabilities Index | 118.35 | 156.44 |
| ARC-AGI-2 | — | 67.1% |
| SimpleBench | — | 75.9% |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 80% |
| ARC-AGI-1 | — | 87.5% |
| CritPt | — | 19.7% |
| Chess Puzzles | — | 40% |
| EBR-Bench | — | 30.5% |
| Mystery Game Puzzles | — | 34% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- Grok 4.6: 67.0 (#24)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 99.2% |
| LMArena Math | 1188 | 1423 |
| FrontierMath (Tiers 1-3) | — | 66% |
| FrontierMath Tier 4 | — | 31.7% |
| ProofBench | — | 51% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- Grok 4.6: 63.3 (#20)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| GPQA Diamond | 36.3% | 94% |
| LMArena Expert | 1148 | 1467 |
| SimpleQA Verified | — | 49.3% |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- Grok 4.6: 43.6 (#23)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Vision | 950 | 1263 |
| Blueprint-Bench 2 | — | 33.2% |
| Furniture Assembly | — | 40% |
| LMArena Document | — | 1452 |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- Grok 4.6: 53.0 (#74)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Non-English | 1178 | 1420 |
| LMArena Chinese | 1155 | 1480 |
| LMArena French | 1195 | 1461 |
| LMArena German | 1174 | 1431 |
| LMArena Japanese | 1102 | 1376 |
| LMArena Korean | 1109 | 1397 |
| LMArena Russian | 1204 | 1422 |
| LMArena Spanish | 1166 | 1404 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- Grok 4.6: 75.4 (#63)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1431 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- Grok 4.6: 44.5 (#66)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Longer Query | 1190 | 1454 |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- Grok 4.6: 62.3 (#80)

| Benchmark | Claude 3 Haiku | Grok 4.6 |
|---|---|---|
| LMArena Text | 1195 | 1428 |
| LMArena Creative Writing | 1157 | 1428 |
| LMArena Multi-Turn | 1190 | 1425 |
| EQ-Bench Creative Writing | 717 | — |

## FAQ

### Is Claude 3 Haiku better than Grok 4.6?

Grok 4.6 is the stronger model overall, scoring 56.9 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or Grok 4.6 better for coding?

Grok 4.6 scores higher on coding benchmarks: 58.5 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Grok 4.6 share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Grok 4.6 has 49.