Model comparison
Llama 4 Maverick vs Mistral Small
Mistral Small is the stronger model overall, scoring 33.1 to 30.6 on the Noometry Index.
Last verified . 31 shared benchmarks.
Meta
Mistral AISummary
- They share 31 benchmarks with published results for both. Llama 4 Maverick scores higher in 4 categories and Mistral Small in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Small leads 52.6 to 39.0.
- The biggest single-benchmark swing is MATH Level 5: 73% for Llama 4 Maverick and 46.8% for Mistral Small.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- Mistral Small accepts more context: 256K tokens versus 128K.
Side by side
| Llama 4 Maverick | Mistral Small | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 30.6 | 33.1 |
| Released | 2025-04-05 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 128K | 256K |
| Max output | 4K | 256K |
| Input $ / M tokens | $0.19 | $0.15 |
| Output $ / M tokens | $0.65 | $0.60 |
| Results tracked | 54 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small leads
Llama 4 Maverick: 26.3 (#318), Mistral Small: 33.5 (#242)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| SciCode | 33.1% | 26.5% |
| BigCodeBench Instruct | 49.7% | 36.1% |
| LMArena Coding | 1302 | 1363 |
| BigCodeBench Complete | 61.4% | 46.6% |
| ALE-Bench | 172.97 | 497.62 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| WeirdML | 24.5% | — |
| LiveBench Coding | — | 36.2% |
Agentic & Tool Use Too close to call
Llama 4 Maverick: 27.9 (#86), Mistral Small: 27.9 (#87)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | 37.1% |
Reasoning Mistral Small leads
Llama 4 Maverick: 10.1 (#337), Mistral Small: 19.3 (#238)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| Kagi LLM Benchmark | 55.9% | 37.8% |
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1281 | 1336 |
| DTBench | 61.9% | 70.9% |
| LMCA | 15.9% | 20.6% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| EnigmaEval | 0.6% | — |
| LiveBench Reasoning | — | 44.8% |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
| LiveBench | — | 44% |
Math Llama 4 Maverick leads
Llama 4 Maverick: 24.6 (#256), Mistral Small: 15.4 (#287)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 20.6% | 5.8% |
| LMArena Math | 1299 | 1341 |
| MATH Level 5 | 73% | 46.8% |
| Omni-MATH | 42.2% | — |
| LiveBench Math | — | 39.9% |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.3 (#196), Mistral Small: 30.9 (#214)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| GPQA Diamond | 67% | 47.5% |
| Vectara Hallucination Rate | 8.2% | 5.1% |
| LMArena Expert | 1259 | 1293 |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| GPQA (HELM) | 65% | — |
| MMLU | — | 68.7% |
Multimodal Mistral Small leads
Llama 4 Maverick: 30.9 (#100), Mistral Small: 32.9 (#91)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| LMArena Vision | 1141 | 1142 |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Mistral Small leads
Llama 4 Maverick: 42.4 (#190), Mistral Small: 45.7 (#164)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| LMArena Non-English | 1269 | 1315 |
| LMArena Chinese | 1277 | 1342 |
| LMArena French | 1258 | 1337 |
| LMArena German | 1291 | 1341 |
| LMArena Japanese | 1207 | 1275 |
| LMArena Korean | 1203 | 1260 |
| LMArena Russian | 1286 | 1324 |
| LMArena Spanish | 1293 | 1346 |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.9 (#140), Mistral Small: 66.6 (#204)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1267 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 90.8% | — |
Long Context Mistral Small leads
Llama 4 Maverick: 31.4 (#274), Mistral Small: 40.4 (#151)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1280 | 1327 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Mistral Small leads
Llama 4 Maverick: 39.0 (#247), Mistral Small: 52.6 (#165)
| Benchmark | Llama 4 Maverick | Mistral Small |
|---|---|---|
| LMArena Text | 1287 | 1338 |
| LMArena Creative Writing | 1266 | 1304 |
| LMArena Multi-Turn | 1289 | 1344 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is Llama 4 Maverick better than Mistral Small?
Mistral Small is the stronger model overall, scoring 33.1 to 30.6 on the Noometry Index.
Which is cheaper, Llama 4 Maverick or Mistral Small?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is Llama 4 Maverick or Mistral Small better for coding?
Mistral Small scores higher on coding benchmarks: 33.5 versus 26.3 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 256K tokens against 128K.
How many benchmarks do Llama 4 Maverick and Mistral Small share?
31 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Mistral Small has 39.