# Claude 3 Haiku vs Mixtral 8x22B

> Mixtral 8x22B is the stronger model overall, scoring 26.7 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-mixtral-8x22b
- Last updated: 2026-10-10
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. Claude 3 Haiku scores higher in 5 categories and Mixtral 8x22B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Mixtral 8x22B leads 21.2 to 9.2.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 24.2% for Mixtral 8x22B.
- Mixtral 8x22B has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 25.7 | 26.7 |
| Rank | 333 | 330 |
| Context | — | 64K |
| Input $/M | — | $2 |
| Output $/M | — | $6 |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- Mixtral 8x22B: 23.8 (#324)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| WeirdML | 9.8% | 3.2% |
| BigCodeBench Instruct | 39.4% | 40.6% |
| LMArena Coding | 1199 | 1166 |
| BigCodeBench Complete | 50.1% | 50.2% |
| HumanEval+ | 68.9% | 72% |
| MBPP+ | 68.8% | 64.3% |
| CadEval | 12% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Mixtral 8x22B: 22.8 (#119)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| Cybench | — | 7.5% |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- Mixtral 8x22B: 19.1 (#246)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1150 |
| DTBench | 50.1% | 55.1% |
| Epoch Capabilities Index | 118.3 | 122 |
| ForecastBench | 53.2 | 56.3 |
| Kagi LLM Benchmark | 34.2% | — |
| LMCA | 8.8% | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- Mixtral 8x22B: 21.2 (#271)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Math | 1188 | 1184 |
| MATH Level 5 | 14.9% | 24.2% |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| Omni-MATH | — | 16.3% |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- Mixtral 8x22B: 15.1 (#288)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| GPQA Diamond | 36.3% | 34.1% |
| LMArena Expert | 1148 | 1113 |
| MMLU | 73.8% | 77.8% |
| MMLU-Pro | — | 46% |
| Confabulations | 34.2% | — |
| GPQA (HELM) | — | 33.4% |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- Mixtral 8x22B: —

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- Mixtral 8x22B: 32.9 (#250)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Non-English | 1178 | 1128 |
| LMArena Chinese | 1155 | 1116 |
| LMArena French | 1194 | 1165 |
| LMArena German | 1174 | 1141 |
| LMArena Japanese | 1102 | 1037 |
| LMArena Korean | 1108 | 1056 |
| LMArena Russian | 1204 | 1158 |
| LMArena Spanish | 1165 | 1150 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- Mixtral 8x22B: 57.9 (#261)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Instruction Following | 1174 | 1147 |
| IFEval | — | 72.4% |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- Mixtral 8x22B: 34.6 (#242)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Longer Query | 1191 | 1144 |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- Mixtral 8x22B: 37.1 (#257)

| Benchmark | Claude 3 Haiku | Mixtral 8x22B |
|---|---|---|
| LMArena Text | 1195 | 1162 |
| LMArena Creative Writing | 1157 | 1141 |
| LMArena Multi-Turn | 1190 | 1130 |
| EQ-Bench Creative Writing | 717 | — |
| WildBench | — | 71.1% |

## FAQ

### Is Claude 3 Haiku better than Mixtral 8x22B?

Mixtral 8x22B is the stronger model overall, scoring 26.7 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or Mixtral 8x22B better for coding?

Claude 3 Haiku scores higher on coding benchmarks: 26.1 versus 23.8 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Mixtral 8x22B share?

28 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Mixtral 8x22B has 34.