Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
MiMo-V2-Omni
~75
Winner · 1/8 categoriesSarvam 105B
60
0/8 categoriesMiMo-V2-Omni· Sarvam 105B
Pick MiMo-V2-Omni if you want the stronger benchmark profile. Sarvam 105B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.
MiMo-V2-Omni is clearly ahead on the aggregate, 75 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
MiMo-V2-Omni's sharpest advantage is in coding, where it averages 74.8 against 45. The single biggest benchmark swing on the page is SWE-bench Verified, 74.8% to 45%.
MiMo-V2-Omni gives you the larger context window at 262K, 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 | MiMo-V2-Omni | Sarvam 105B |
|---|---|---|
| Agentic | ||
| Claw-Eval | 56.7% | — |
| BrowseComp | — | 49.5% |
| CodingMiMo-V2-Omni wins | ||
| SWE-bench Verified | 74.8% | 45% |
| LiveCodeBench v6 | — | 71.7% |
| Multimodal & Grounded | ||
| MMMU-Pro | 76.8% | — |
| Reasoning | ||
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| Knowledge | ||
| MMLU | — | 90.6% |
| MMLU-Pro | — | 81.7% |
| Instruction Following | ||
| IFEval | — | 84.8% |
| Multilingual | ||
| Coming soon | ||
| Mathematics | ||
| MATH-500 | — | 98.6% |
| AIME 2025 | — | 88.3% |
| HMMT Feb 2025 | — | 85.8% |
| HMMT Nov 2025 | — | 85.8% |
MiMo-V2-Omni is ahead overall, 75 to 60. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 74.8% and 45%.
MiMo-V2-Omni has the edge for coding in this comparison, averaging 74.8 versus 45. Inside this category, SWE-bench Verified 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.