Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Mistral Small 4 (Reasoning)
~64
Winner · 1/8 categoriesSarvam 105B
60
2/8 categoriesMistral Small 4 (Reasoning)· Sarvam 105B
Pick Mistral Small 4 (Reasoning) if you want the stronger benchmark profile. Sarvam 105B only becomes the better choice if mathematics is the priority.
Mistral Small 4 (Reasoning) is clearly ahead on the aggregate, 64 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Mistral Small 4 (Reasoning)'s sharpest advantage is in coding, where it averages 63.6 against 45. The single biggest benchmark swing on the page is AIME 2025, 83.8% to 88.3%. Sarvam 105B does hit back in mathematics, so the answer changes if that is the part of the workload you care about most.
Mistral Small 4 (Reasoning) gives you the larger context window at 256K, compared with 128K for Sarvam 105B.
BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.
Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.
| Benchmark | Mistral Small 4 (Reasoning) | Sarvam 105B |
|---|---|---|
| Agentic | ||
| BrowseComp | — | 49.5% |
| CodingMistral Small 4 (Reasoning) wins | ||
| LiveCodeBench | 63.6% | — |
| LiveCodeBench v6 | — | 71.7% |
| SWE-bench Verified | — | 45% |
| Multimodal & Grounded | ||
| MMMU-Pro | 60% | — |
| Reasoning | ||
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| KnowledgeSarvam 105B wins | ||
| GPQA | 71.2% | — |
| MMLU-Pro | 78% | 81.7% |
| MMLU | — | 90.6% |
| Instruction Following | ||
| IFEval | — | 84.8% |
| Multilingual | ||
| Coming soon | ||
| MathematicsSarvam 105B wins | ||
| AIME 2025 | 83.8% | 88.3% |
| MATH-500 | — | 98.6% |
| HMMT Feb 2025 | — | 85.8% |
| HMMT Nov 2025 | — | 85.8% |
Mistral Small 4 (Reasoning) is ahead overall, 64 to 60. The biggest single separator in this matchup is AIME 2025, where the scores are 83.8% and 88.3%.
Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 75.6. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
Mistral Small 4 (Reasoning) has the edge for coding in this comparison, averaging 63.6 versus 45. Sarvam 105B stays close enough that the answer can still flip depending on your workload.
Sarvam 105B has the edge for math in this comparison, averaging 92.3 versus 83.8. Inside this category, AIME 2025 is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.