# Claude 3 Haiku vs Mistral Large 3

> Mistral Large 3 is the stronger model overall, scoring 39.1 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-mistral-large-3
- Last updated: 2026-10-12
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Claude 3 Haiku scores higher in 1 category and Mistral Large 3 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Large 3 leads 60.0 to 29.7.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 34.2% for Claude 3 Haiku and 50.9% for Mistral Large 3.
- Mistral Large 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 25.9 | 39.1 |
| Rank | 340 | 176 |
| Context | — | 262K |
| Input $/M | — | $0.25 |
| Output $/M | — | $0.75 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- Mistral Large 3: 34.4 (#237)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Coding | 1199 | 1448 |
| LMArena WebDev | — | 1230 |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- Mistral Large 3: 15.2 (#319)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 50.9% |
| LMArena Hard Prompts | 1174 | 1429 |
| NYT Connections (extended) | — | 7.5% |
| Thematic Generalization | — | 23% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 118.35 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- Mistral Large 3: 38.7 (#129)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Math | 1188 | 1414 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- Mistral Large 3: 36.0 (#177)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Expert | 1148 | 1421 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 14.5% |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- Mistral Large 3: 38.2 (#66)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Vision | 950 | 1221 |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- Mistral Large 3: 52.5 (#84)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Non-English | 1178 | 1413 |
| LMArena Chinese | 1155 | 1447 |
| LMArena French | 1195 | 1455 |
| LMArena German | 1174 | 1437 |
| LMArena Japanese | 1102 | 1394 |
| LMArena Korean | 1109 | 1384 |
| LMArena Russian | 1204 | 1411 |
| LMArena Spanish | 1166 | 1440 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- Mistral Large 3: 74.0 (#108)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Instruction Following | 1173 | 1403 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- Mistral Large 3: 43.1 (#105)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Longer Query | 1190 | 1413 |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- Mistral Large 3: 60.0 (#101)

| Benchmark | Claude 3 Haiku | Mistral Large 3 |
|---|---|---|
| LMArena Text | 1195 | 1428 |
| LMArena Creative Writing | 1157 | 1386 |
| EQ-Bench Creative Writing | 717 | 1412 |
| LMArena Multi-Turn | 1190 | 1429 |

## FAQ

### Is Claude 3 Haiku better than Mistral Large 3?

Mistral Large 3 is the stronger model overall, scoring 39.1 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or Mistral Large 3 better for coding?

Mistral Large 3 scores higher on coding benchmarks: 34.4 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Mistral Large 3 share?

20 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Mistral Large 3 has 24.