# Claude Haiku 4.5 vs Mistral Medium

> Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 36.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-haiku-4-5-vs-mistral-medium
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 5 categories and Mistral Medium in 5 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 44.9 to 28.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 32.2% for Mistral Medium.
- Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 200K.
- Mistral Medium has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 39.5 | 36.3 |
| Rank | 165 | 218 |
| Context | 200K | 262K |
| Input $/M | $1 | $1.50 |
| Output $/M | $5 | $7.50 |
| Weights | Proprietary | Open |

## Coding

- Claude Haiku 4.5: 44.0 (#78)
- Mistral Medium: 34.2 (#243)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| SciCode | 43.3% | 40.2% |
| WeirdML | 45.4% | 43.7% |
| LMArena Coding | 1453 | 1434 |
| ALE-Bench | 653.48 | 763.98 |
| FrontierCode | — | 8% |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |

## Agentic & Tool Use

- Claude Haiku 4.5: 33.6 (#52)
- Mistral Medium: 28.3 (#90)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 68.7% | 37.7% |
| Terminal-Bench | 35.5% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |

## Reasoning

- Claude Haiku 4.5: 15.1 (#320)
- Mistral Medium: 24.0 (#167)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1420 | 1426 |
| DTBench | 73.6% | 75.5% |
| LMCA | 30.9% | 26.1% |
| ARC-AGI-2 | 4% | — |
| Kagi LLM Benchmark | — | 50% |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| Chess Puzzles | 8% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 142.41 | — |
| ForecastBench | 61.4 | — |

## Math

- Claude Haiku 4.5: 44.9 (#78)
- Mistral Medium: 28.1 (#245)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 32.2% |
| LMArena Math | 1396 | 1408 |
| MATH Level 5 | 96.4% | 81.6% |
| FrontierMath (Feb 2025 set) | 5.9% | 0.3% |
| ProofBench | — | 9% |
| Omni-MATH | 56.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |

## Knowledge

- Claude Haiku 4.5: 37.7 (#153)
- Mistral Medium: 25.0 (#265)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| GPQA Diamond | 71.2% | 59.5% |
| Vectara Hallucination Rate | 9.8% | 22.7% |
| LMArena Expert | 1442 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| SimpleQA Verified | 13.2% | — |
| MMLU-Pro | 77.7% | — |
| GPQA (HELM) | 60.5% | — |

## Multimodal

- Claude Haiku 4.5: 26.8 (#118)
- Mistral Medium: 35.3 (#88)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |

## Multilingual

- Claude Haiku 4.5: 49.9 (#129)
- Mistral Medium: 52.1 (#91)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1377 | 1408 |
| LMArena Chinese | 1417 | 1447 |
| LMArena French | 1408 | 1459 |
| LMArena German | 1375 | 1432 |
| LMArena Japanese | 1339 | 1378 |
| LMArena Korean | 1347 | 1380 |
| LMArena Russian | 1381 | 1411 |
| LMArena Spanish | 1420 | 1433 |

## Instruction Following

- Claude Haiku 4.5: 71.4 (#149)
- Mistral Medium: 73.7 (#116)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1414 | 1398 |
| IFEval | 80.1% | — |

## Long Context

- Claude Haiku 4.5: 43.6 (#92)
- Mistral Medium: 42.9 (#114)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1427 | 1406 |

## Writing & Preference

- Claude Haiku 4.5: 57.9 (#123)
- Mistral Medium: 60.0 (#103)

| Benchmark | Claude Haiku 4.5 | Mistral Medium |
|---|---|---|
| LMArena Text | 1396 | 1424 |
| LMArena Creative Writing | 1372 | 1391 |
| LMArena Multi-Turn | 1409 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| WildBench | 83.9% | — |
| EQ-Bench 4 | 1064 | — |

## FAQ

### Is Claude Haiku 4.5 better than Mistral Medium?

Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 36.3 on the Noometry Index.

### Which is cheaper, Claude Haiku 4.5 or Mistral Medium?

Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; Mistral Medium lists at $1.50 and $7.50.

### Is Claude Haiku 4.5 or Mistral Medium better for coding?

Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

Mistral Medium does, with 262K tokens against 200K.

### How many benchmarks do Claude Haiku 4.5 and Mistral Medium share?

29 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Mistral Medium has 36.