# Claude 3.5 Haiku vs Grok 3

> Grok 3 is the stronger model overall, scoring 39.9 to 29.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-5-haiku-vs-grok-3
- Last updated: 2026-10-12
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and Grok 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 3 leads 46.2 to 18.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 55.6% for Grok 3.

## Snapshot

| | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 29.2 | 39.9 |
| Rank | 315 | 157 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3.5 Haiku: 32.9 (#265)
- Grok 3: 41.9 (#115)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| Aider Polyglot | 28% | 53.3% |
| WeirdML | 30.7% | 37.2% |
| LMArena Coding | 1286 | 1432 |
| SciCode | 27.4% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |

## Agentic & Tool Use

- Claude 3.5 Haiku: 28.0 (#95)
- Grok 3: 30.5 (#76)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| BALROG | 19.3% | 29.5% |

## Reasoning

- Claude 3.5 Haiku: 17.7 (#290)
- Grok 3: 13.7 (#333)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| LMArena Hard Prompts | 1251 | 1434 |
| Epoch Capabilities Index | 127.15 | 138.33 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| Kagi LLM Benchmark | — | 61.3% |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 0% | — |
| LiveBench Reasoning | 28.1% | — |
| DTBench | 56.7% | — |
| LiveBench Data Analysis | 48.5% | — |
| LiveBench | 43.5% | — |

## Math

- Claude 3.5 Haiku: 14.7 (#300)
- Grok 3: 38.0 (#145)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 55.6% |
| Omni-MATH | 22.4% | 46.4% |
| LMArena Math | 1244 | 1391 |
| MATH Level 5 | 46.4% | 88.7% |
| FrontierMath (Feb 2025 set) | 0.3% | 3.8% |
| LiveBench Math | 35.5% | — |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Claude 3.5 Haiku: 18.7 (#281)
- Grok 3: 46.2 (#82)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| GPQA Diamond | 38.1% | 75.8% |
| MMLU-Pro | 60.5% | 78.8% |
| Confabulations | 36.7% | 14.2% |
| GPQA (HELM) | 36.3% | 65% |
| LMArena Expert | 1208 | 1421 |
| Vectara Hallucination Rate | — | 5.8% |
| MMLU | 74.3% | — |

## Multimodal

- Claude 3.5 Haiku: 26.8 (#117)
- Grok 3: —

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| LMArena Vision | 1092 | — |
| GeoBench | 34% | — |

## Multilingual

- Claude 3.5 Haiku: 40.0 (#218)
- Grok 3: 52.3 (#87)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| LMArena Non-English | 1238 | 1410 |
| LMArena Chinese | 1229 | 1448 |
| LMArena French | 1264 | 1460 |
| LMArena German | 1237 | 1431 |
| LMArena Japanese | 1175 | 1387 |
| LMArena Korean | 1173 | 1373 |
| LMArena Russian | 1253 | 1416 |
| LMArena Spanish | 1261 | 1417 |

## Instruction Following

- Claude 3.5 Haiku: 62.9 (#234)
- Grok 3: 75.0 (#73)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| IFEval | 79.2% | 88.4% |
| LMArena Instruction Following | 1241 | 1409 |
| LiveBench Instruction Following | 61.9% | — |

## Long Context

- Claude 3.5 Haiku: 38.3 (#200)
- Grok 3: 38.7 (#192)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1261 | 1439 |
| Fiction.LiveBench | — | 58.3% |

## Writing & Preference

- Claude 3.5 Haiku: 42.7 (#234)
- Grok 3: 55.8 (#141)

| Benchmark | Claude 3.5 Haiku | Grok 3 |
|---|---|---|
| LMArena Text | 1255 | 1426 |
| LMArena Creative Writing | 1233 | 1414 |
| Short-Story Creative Writing | 73.5% | 76.4% |
| EQ-Bench Creative Writing | 1146 | 1186 |
| WildBench | 76% | 84.9% |
| LMArena Multi-Turn | 1265 | 1425 |
| LiveBench Language | 35.4% | — |

## FAQ

### Is Claude 3.5 Haiku better than Grok 3?

Grok 3 is the stronger model overall, scoring 39.9 to 29.2 on the Noometry Index.

### Is Claude 3.5 Haiku or Grok 3 better for coding?

Grok 3 scores higher on coding benchmarks: 41.9 versus 32.9 in the Noometry coding category.

### How many benchmarks do Claude 3.5 Haiku and Grok 3 share?

33 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Grok 3 has 40.