Model comparison
Claude 3 Haiku vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.0 to 25.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
Mistral AISummary
- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Mistral Medium in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.2 to 29.8.
- The biggest single-benchmark swing is MATH Level 5: 14.9% for Claude 3 Haiku and 81.6% for Mistral Medium.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Mistral Medium | |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 25.7 | 36.0 |
| Released | 2024-03-07 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $1.50 |
| Output $ / M tokens | — | $7.50 |
| Results tracked | 37 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Claude 3 Haiku: 26.1 (#320), Mistral Medium: 33.8 (#241)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| WeirdML | 9.8% | 43.7% |
| LMArena Coding | 1199 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| ALE-Bench | — | 763.98 |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Agentic & Tool Use Not comparable
Claude 3 Haiku: —, Mistral Medium: 28.1 (#84)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.7% |
Reasoning Mistral Medium leads
Claude 3 Haiku: 15.7 (#303), Mistral Medium: 23.3 (#165)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| Kagi LLM Benchmark | 34.2% | 50% |
| LMArena Hard Prompts | 1174 | 1426 |
| DTBench | 50.1% | 75.5% |
| LMCA | 8.8% | 26.1% |
| CritPt | — | 0% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 118.3 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Mistral Medium leads
Claude 3 Haiku: 9.2 (#313), Mistral Medium: 27.2 (#238)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 32.2% |
| LMArena Math | 1188 | 1409 |
| MATH Level 5 | 14.9% | 81.6% |
| ProofBench | — | 9% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Mistral Medium leads
Claude 3 Haiku: 17.3 (#280), Mistral Medium: 24.9 (#260)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| GPQA Diamond | 36.3% | 59.5% |
| LMArena Expert | 1148 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| Confabulations | 34.2% | — |
| Vectara Hallucination Rate | — | 22.7% |
| MMLU | 73.8% | — |
Multimodal Mistral Medium leads
Claude 3 Haiku: 23.1 (#123), Mistral Medium: 34.6 (#82)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| LMArena Vision | 950 | 1172 |
| ScienceQA | 72% | — |
Multilingual Mistral Medium leads
Claude 3 Haiku: 36.1 (#238), Mistral Medium: 52.3 (#86)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1178 | 1408 |
| LMArena Chinese | 1155 | 1448 |
| LMArena French | 1194 | 1460 |
| LMArena German | 1174 | 1433 |
| LMArena Japanese | 1102 | 1378 |
| LMArena Korean | 1108 | 1382 |
| LMArena Russian | 1204 | 1411 |
| LMArena Spanish | 1165 | 1433 |
Instruction Following Mistral Medium leads
Claude 3 Haiku: 61.5 (#242), Mistral Medium: 73.9 (#111)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1174 | 1398 |
Long Context Mistral Medium leads
Claude 3 Haiku: 36.1 (#232), Mistral Medium: 42.9 (#108)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1191 | 1406 |
Writing & Preference Mistral Medium leads
Claude 3 Haiku: 29.8 (#286), Mistral Medium: 60.2 (#97)
| Benchmark | Claude 3 Haiku | Mistral Medium |
|---|---|---|
| LMArena Text | 1195 | 1424 |
| LMArena Creative Writing | 1157 | 1391 |
| LMArena Multi-Turn | 1190 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.0 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 33.8 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Mistral Medium share?
25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Mistral Medium has 36.