Model comparison
Claude 3.5 Haiku vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 29.2 on the Noometry Index.
Last verified . 27 shared benchmarks.
Mistral AISummary
- They share 27 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and Mistral Medium in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 42.7.
- The biggest single-benchmark swing is MATH Level 5: 46.4% for Claude 3.5 Haiku and 81.6% for Mistral Medium.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | Mistral Medium | |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 29.2 | 36.3 |
| Released | 2024-10-22 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $1.50 |
| Output $ / M tokens | — | $7.50 |
| Results tracked | 49 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Claude 3.5 Haiku: 32.9 (#265), Mistral Medium: 34.2 (#243)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| SciCode | 27.4% | 40.2% |
| WeirdML | 30.7% | 43.7% |
| LMArena Coding | 1286 | 1434 |
| FrontierCode | — | 8% |
| Aider Polyglot | 28% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Too close to call
Claude 3.5 Haiku: 28.0 (#95), Mistral Medium: 28.3 (#90)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
| BALROG | 19.3% | — |
Reasoning Mistral Medium leads
Claude 3.5 Haiku: 17.7 (#290), Mistral Medium: 24.0 (#167)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1251 | 1426 |
| DTBench | 56.7% | 75.5% |
| Kagi LLM Benchmark | — | 50% |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 127.15 | — |
| LiveBench | 43.5% | — |
Math Mistral Medium leads
Claude 3.5 Haiku: 14.7 (#300), Mistral Medium: 28.1 (#245)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 32.2% |
| LMArena Math | 1244 | 1408 |
| MATH Level 5 | 46.4% | 81.6% |
| FrontierMath (Feb 2025 set) | 0.3% | 0.3% |
| ProofBench | — | 9% |
| Omni-MATH | 22.4% | — |
| LiveBench Math | 35.5% | — |
Knowledge Mistral Medium leads
Claude 3.5 Haiku: 18.7 (#281), Mistral Medium: 25.0 (#265)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| GPQA Diamond | 38.1% | 59.5% |
| LMArena Expert | 1208 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 22.7% |
| GPQA (HELM) | 36.3% | — |
| MMLU | 74.3% | — |
Multimodal Mistral Medium leads
Claude 3.5 Haiku: 26.8 (#117), Mistral Medium: 35.3 (#88)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| LMArena Vision | 1092 | 1172 |
| GeoBench | 34% | — |
Multilingual Mistral Medium leads
Claude 3.5 Haiku: 40.0 (#218), Mistral Medium: 52.1 (#91)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1238 | 1408 |
| LMArena Chinese | 1229 | 1447 |
| LMArena French | 1264 | 1459 |
| LMArena German | 1237 | 1432 |
| LMArena Japanese | 1175 | 1378 |
| LMArena Korean | 1173 | 1380 |
| LMArena Russian | 1253 | 1411 |
| LMArena Spanish | 1261 | 1433 |
Instruction Following Mistral Medium leads
Claude 3.5 Haiku: 62.9 (#234), Mistral Medium: 73.7 (#116)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1241 | 1398 |
| LiveBench Instruction Following | 61.9% | — |
| IFEval | 79.2% | — |
Long Context Mistral Medium leads
Claude 3.5 Haiku: 38.3 (#200), Mistral Medium: 42.9 (#114)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1261 | 1406 |
Writing & Preference Mistral Medium leads
Claude 3.5 Haiku: 42.7 (#234), Mistral Medium: 60.0 (#103)
| Benchmark | Claude 3.5 Haiku | Mistral Medium |
|---|---|---|
| LMArena Text | 1255 | 1424 |
| LMArena Creative Writing | 1233 | 1391 |
| Short-Story Creative Writing | 73.5% | 77.3% |
| LMArena Multi-Turn | 1265 | 1418 |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Mistral Medium share?
27 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Mistral Medium has 36.