# Claude 3 Haiku vs Grok 4

> Grok 4 is the stronger model overall, scoring 47.8 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-grok-4
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Grok 4 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 4 leads 46.5 to 9.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 84% for Grok 4.

## Snapshot

| | Claude 3 Haiku | Grok 4 |
|---|---|---|
| Provider | Anthropic | xAI |
| Noometry Index | 25.7 | 47.8 |
| Rank | 333 | 49 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- Grok 4: 49.9 (#44)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| WeirdML | 9.8% | 45.7% |
| LMArena Coding | 1199 | 1408 |
| Aider Polyglot | — | 79.6% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Grok 4: 32.1 (#62)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| Terminal-Bench | — | 27.2% |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 21.1% |
| Cybench | — | 43% |
| DeepResearch Bench | — | 47.3% |
| BALROG | — | 43.6% |
| LMArena Search | — | 1142 |
| METR Time Horizons | — | 66.6% |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- Grok 4: 36.9 (#59)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 73.6% |
| LMArena Hard Prompts | 1174 | 1409 |
| Epoch Capabilities Index | 118.3 | 146.46 |
| ForecastBench | 53.2 | 60.9 |
| ARC-AGI-2 | — | 16% |
| SimpleBench | — | 60.5% |
| ARC-AGI-1 | — | 66.7% |
| Chess Puzzles | — | 28% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- Grok 4: 46.5 (#58)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 84% |
| LMArena Math | 1188 | 1422 |
| Omni-MATH | — | 60.3% |
| MATH Level 5 | 14.9% | — |
| FrontierMath (Feb 2025 set) | — | 19.7% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- Grok 4: 53.7 (#50)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| GPQA Diamond | 36.3% | 87% |
| Confabulations | 34.2% | 12.4% |
| LMArena Expert | 1148 | 1417 |
| MMLU-Pro | — | 85.1% |
| GPQA (HELM) | — | 72.7% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- Grok 4: 33.1 (#89)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| LMArena Vision | 950 | 1211 |
| GeoBench | — | 45% |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- Grok 4: 52.0 (#98)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| LMArena Non-English | 1178 | 1403 |
| LMArena Chinese | 1155 | 1428 |
| LMArena French | 1194 | 1418 |
| LMArena German | 1174 | 1429 |
| LMArena Japanese | 1102 | 1396 |
| LMArena Korean | 1108 | 1376 |
| LMArena Russian | 1204 | 1409 |
| LMArena Spanish | 1165 | 1419 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- Grok 4: 79.4 (#4)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| LMArena Instruction Following | 1174 | 1387 |
| IFEval | — | 94.9% |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- Grok 4: 63.1 (#4)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| LMArena Longer Query | 1191 | 1409 |
| Fiction.LiveBench | — | 94.4% |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- Grok 4: 58.7 (#111)

| Benchmark | Claude 3 Haiku | Grok 4 |
|---|---|---|
| LMArena Text | 1195 | 1412 |
| LMArena Creative Writing | 1157 | 1397 |
| LMArena Multi-Turn | 1190 | 1416 |
| Short-Story Creative Writing | — | 76.9% |
| EQ-Bench Creative Writing | 717 | — |
| WildBench | — | 79.7% |

## FAQ

### Is Claude 3 Haiku better than Grok 4?

Grok 4 is the stronger model overall, scoring 47.8 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or Grok 4 better for coding?

Grok 4 scores higher on coding benchmarks: 49.9 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Grok 4 share?

25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Grok 4 has 48.