Model comparison
Claude Sonnet 4 vs Mistral Small
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 33.4 on the Noometry Index. Mistral Small costs 23× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Mistral AISummary
- They share 28 benchmarks with published results for both. Claude Sonnet 4 scores higher in 8 categories and Mistral Small in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4 leads 43.3 to 16.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 71.1% for Claude Sonnet 4 and 5.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.
- Mistral Small accepts more context: 262K tokens versus 200K.
- Mistral Small has downloadable open weights; the other is API-only.
Side by side
| Claude Sonnet 4 | Mistral Small | |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 40.8 | 33.4 |
| Released | 2025-05-22 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 200K | 262K |
| Max output | 64K | 256K |
| Input $ / M tokens | $3 | $0.15 |
| Output $ / M tokens | $15 | $0.60 |
| Results tracked | 58 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4 leads
Claude Sonnet 4: 43.5 (#88), Mistral Small: 34.0 (#247)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| SciCode | 40% | 26.5% |
| LMArena Coding | 1414 | 1362 |
| ALE-Bench | 655.35 | 497.62 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| BigCodeBench Instruct | — | 36.1% |
| LiveBench Coding | — | 36.2% |
| BigCodeBench Complete | — | 46.6% |
Agentic & Tool Use Claude Sonnet 4 leads
Claude Sonnet 4: 38.5 (#31), Mistral Small: 28.1 (#93)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.1% |
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning Claude Sonnet 4 leads
Claude Sonnet 4: 22.9 (#187), Mistral Small: 19.8 (#250)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 73% | 37.8% |
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1372 | 1335 |
| DTBench | 77.1% | 70.9% |
| LMCA | 29% | 20.6% |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| ARC-AGI-1 | 40% | — |
| EnigmaEval | 3.1% | — |
| LiveBench Reasoning | — | 44.8% |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 141.69 | — |
| ForecastBench | 60.2 | — |
| LiveBench | — | 44% |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), Mistral Small: 16.4 (#293)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 5.8% |
| LMArena Math | 1375 | 1341 |
| MATH Level 5 | 84.4% | 46.8% |
| Omni-MATH | 60.2% | — |
| LiveBench Math | — | 39.9% |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Claude Sonnet 4 leads
Claude Sonnet 4: 41.8 (#108), Mistral Small: 31.0 (#222)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| GPQA Diamond | 79.2% | 47.5% |
| Vectara Hallucination Rate | 10.3% | 5.1% |
| LMArena Expert | 1372 | 1291 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| GPQA (HELM) | 70.6% | — |
| MMLU | — | 68.7% |
Multimodal Mistral Small leads
Claude Sonnet 4: 26.2 (#121), Mistral Small: 33.5 (#96)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| LMArena Vision | 1191 | 1142 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| MindCube | 44.8% | — |
Multilingual Claude Sonnet 4 leads
Claude Sonnet 4: 46.7 (#156), Mistral Small: 45.5 (#169)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| LMArena Non-English | 1333 | 1315 |
| LMArena Chinese | 1350 | 1340 |
| LMArena French | 1363 | 1337 |
| LMArena German | 1331 | 1340 |
| LMArena Japanese | 1302 | 1275 |
| LMArena Korean | 1291 | 1259 |
| LMArena Russian | 1355 | 1324 |
| LMArena Spanish | 1357 | 1346 |
Instruction Following Claude Sonnet 4 leads
Claude Sonnet 4: 71.7 (#145), Mistral Small: 66.4 (#209)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1376 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 84% | — |
Long Context Mistral Small leads
Claude Sonnet 4: 33.7 (#259), Mistral Small: 40.4 (#156)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1398 | 1327 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference Claude Sonnet 4 leads
Claude Sonnet 4: 57.1 (#132), Mistral Small: 52.5 (#171)
| Benchmark | Claude Sonnet 4 | Mistral Small |
|---|---|---|
| LMArena Text | 1351 | 1338 |
| LMArena Creative Writing | 1345 | 1305 |
| LMArena Multi-Turn | 1376 | 1344 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is Claude Sonnet 4 better than Mistral Small?
Claude Sonnet 4 is the stronger model overall, scoring 40.8 to 33.4 on the Noometry Index. Mistral Small costs 23× less per token, which makes it the better buy when Claude Sonnet 4's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4 or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Claude Sonnet 4 lists at $3 and $15.
Is Claude Sonnet 4 or Mistral Small better for coding?
Claude Sonnet 4 scores higher on coding benchmarks: 43.5 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 262K tokens against 200K.
How many benchmarks do Claude Sonnet 4 and Mistral Small share?
28 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and Mistral Small has 39.