Model comparison
Claude 3.5 Haiku vs Mistral Small
Mistral Small is the stronger model overall, scoring 33.1 to 28.9 on the Noometry Index.
Last verified . 34 shared benchmarks.
Mistral AISummary
- They share 34 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 0 categories and Mistral Small in 10 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Small leads 30.9 to 18.7.
- The biggest single-benchmark swing is LiveBench Reasoning: 28.1% for Claude 3.5 Haiku and 44.8% for Mistral Small.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Haiku | Mistral Small | |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 28.9 | 33.1 |
| Released | 2024-10-22 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | — | 256K |
| Max output | — | 256K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.60 |
| Results tracked | 49 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude 3.5 Haiku: 32.6 (#258), Mistral Small: 33.5 (#242)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| SciCode | 27.4% | 26.5% |
| BigCodeBench Instruct | 46.1% | 36.1% |
| LiveBench Coding | 51.4% | 36.2% |
| LMArena Coding | 1287 | 1363 |
| BigCodeBench Complete | 59% | 46.6% |
| Aider Polyglot | 28% | — |
| WeirdML | 30.7% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 497.62 |
Agentic & Tool Use Too close to call
Claude 3.5 Haiku: 27.7 (#90), Mistral Small: 27.9 (#87)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
| BALROG | 19.3% | — |
Reasoning Mistral Small leads
Claude 3.5 Haiku: 17.1 (#286), Mistral Small: 19.3 (#238)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| CritPt | 0% | 0% |
| LiveBench Reasoning | 28.1% | 44.8% |
| LMArena Hard Prompts | 1251 | 1336 |
| DTBench | 56.7% | 70.9% |
| LiveBench Data Analysis | 48.5% | 53.7% |
| LiveBench | 43.5% | 44% |
| Kagi LLM Benchmark | — | 37.8% |
| LMCA | — | 20.6% |
| Epoch Capabilities Index | 127.15 | — |
Math Mistral Small leads
Claude 3.5 Haiku: 13.7 (#294), Mistral Small: 15.4 (#287)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 5.8% |
| LiveBench Math | 35.5% | 39.9% |
| LMArena Math | 1244 | 1341 |
| MATH Level 5 | 46.4% | 46.8% |
| Omni-MATH | 22.4% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Mistral Small leads
Claude 3.5 Haiku: 18.7 (#276), Mistral Small: 30.9 (#214)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| GPQA Diamond | 38.1% | 47.5% |
| LMArena Expert | 1208 | 1293 |
| MMLU | 74.3% | 68.7% |
| MMLU-Pro | 60.5% | — |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 5.1% |
| GPQA (HELM) | 36.3% | — |
Multimodal Mistral Small leads
Claude 3.5 Haiku: 26.3 (#113), Mistral Small: 32.9 (#91)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| LMArena Vision | 1092 | 1142 |
| GeoBench | 34% | — |
Multilingual Mistral Small leads
Claude 3.5 Haiku: 40.2 (#213), Mistral Small: 45.7 (#164)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| LMArena Non-English | 1238 | 1315 |
| LMArena Chinese | 1229 | 1342 |
| LMArena French | 1262 | 1337 |
| LMArena German | 1237 | 1341 |
| LMArena Japanese | 1175 | 1275 |
| LMArena Korean | 1172 | 1260 |
| LMArena Russian | 1253 | 1324 |
| LMArena Spanish | 1261 | 1346 |
Instruction Following Mistral Small leads
Claude 3.5 Haiku: 63.1 (#229), Mistral Small: 66.6 (#204)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| LiveBench Instruction Following | 61.9% | 63.7% |
| LMArena Instruction Following | 1241 | 1310 |
| IFEval | 79.2% | — |
Long Context Mistral Small leads
Claude 3.5 Haiku: 38.3 (#195), Mistral Small: 40.4 (#151)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1261 | 1327 |
Writing & Preference Mistral Small leads
Claude 3.5 Haiku: 42.9 (#228), Mistral Small: 52.6 (#165)
| Benchmark | Claude 3.5 Haiku | Mistral Small |
|---|---|---|
| LMArena Text | 1255 | 1338 |
| LMArena Creative Writing | 1233 | 1304 |
| LMArena Multi-Turn | 1265 | 1344 |
| LiveBench Language | 35.4% | 30.5% |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| WildBench | 76% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Mistral Small?
Mistral Small is the stronger model overall, scoring 33.1 to 28.9 on the Noometry Index.
Is Claude 3.5 Haiku or Mistral Small better for coding?
They score almost the same on coding (32.6 vs 33.5); test both on your own repository before choosing.
How many benchmarks do Claude 3.5 Haiku and Mistral Small share?
34 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Mistral Small has 39.