Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Kimi K2.5
72
Winner · 3/8 categoriesSarvam 105B
60
2/8 categoriesKimi K2.5· Sarvam 105B
Pick Kimi K2.5 if you want the stronger benchmark profile. Sarvam 105B only becomes the better choice if knowledge is the priority or you want the cheaper token bill.
Kimi K2.5 is clearly ahead on the aggregate, 72 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Kimi K2.5's sharpest advantage is in coding, where it averages 66.7 against 45. The single biggest benchmark swing on the page is SWE-bench Verified, 76.8% to 45%. Sarvam 105B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Kimi K2.5 is also the more expensive model on tokens at $0.50 input / $2.80 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Sarvam 105B. That is roughly Infinityx on output cost alone. Sarvam 105B is the reasoning model in the pair, while Kimi K2.5 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use.
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 | Kimi K2.5 | Sarvam 105B |
|---|---|---|
| AgenticKimi K2.5 wins | ||
| Terminal-Bench 2.0 | 50.8% | — |
| BrowseComp | 60.6% | 49.5% |
| OSWorld-Verified | 63.3% | — |
| Claw-Eval | 52.9% | — |
| QwenClawBench | 54.3% | — |
| QwenWebBench | 1160 | — |
| TAU3-Bench | 65.7% | — |
| VITA-Bench | 36.0% | — |
| DeepPlanning | 14.4% | — |
| Toolathlon | 27.8% | — |
| MCP Atlas | 29.5% | — |
| MCP-Tasks | 59.1% | — |
| WideResearch | 72.7% | — |
| BrowseComp-VL | 42.9% | — |
| OSWorld | 63.3% | — |
| Tau2-Airline | 80.0% | — |
| Tau2-Telecom | 95.9% | — |
| BFCL v4 | 68.3% | — |
| AndroidWorld | 43.1% | — |
| WebVoyager | 84.3% | — |
| CodingKimi K2.5 wins | ||
| HumanEval | 99% | — |
| SWE-bench Verified | 76.8% | 45% |
| SWE-bench Verified* | 70.8% | — |
| LiveCodeBench | 85% | — |
| LiveCodeBench v6 | 85.6% | 71.7% |
| SWE-bench Pro | 53.8% | — |
| SWE Multilingual | 73% | — |
| SWE-Rebench | 58.5% | — |
| React Native Evals | 74.9% | — |
| NL2Repo | 32% | — |
| Multimodal & Grounded | ||
| MMMU | 84.3% | — |
| MMMU-Pro | 78.5% | — |
| OfficeQA Pro | 69% | — |
| RealWorldQA | 81.0% | — |
| OmniDocBench 1.5 | 88.8% | — |
| Video-MME (with subtitle) | 87.4% | — |
| Video-MME (w/o subtitle) | 83.2% | — |
| MathVision | 84.2% | — |
| We-Math | 84.7% | — |
| DynaMath | 84.4% | — |
| MStar | 80.5% | — |
| Design2Code | 91.3% | — |
| Flame-VLM-Code | 88.8% | — |
| Vision2Web | 33.2% | — |
| ImageMining | 24.4% | — |
| MMSearch | 58.7% | — |
| MMSearch-Plus | 25.6% | — |
| SimpleVQA | 71.2% | — |
| ChatCVQA | 77.5% | — |
| MMLongBench-Doc | 58.5% | — |
| CC-OCR | 79.7% | — |
| AI2D_TEST | 90.8% | — |
| CountBench | 94.1% | — |
| RefCOCO (avg) | 87.8% | — |
| VideoMMMU | 86.6% | — |
| MLVU (M-Avg) | 85.0% | — |
| Facts-VLM | 57.8% | — |
| V* | 84.3% | — |
| Reasoning | ||
| MuSR | 72% | — |
| BBH | 81% | — |
| LongBench v2 | 61% | — |
| MRCRv2 | 70% | — |
| AI-Needle | 70% | — |
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| KnowledgeSarvam 105B wins | ||
| MMLU | 77% | 90.6% |
| GPQA | 87.6% | — |
| GPQA-D | 86.9% | — |
| SuperGPQA | 69.2% | — |
| MMLU-Pro | 87.1% | 81.7% |
| MMLU-Pro (Arcee) | 87.1% | — |
| MMLU-Redux | 94.5% | — |
| C-Eval | 94% | — |
| HLE | 30.1% | — |
| FrontierScience | 67% | — |
| SimpleQA | 74% | — |
| Instruction FollowingKimi K2.5 wins | ||
| IFEval | 93.9% | 84.8% |
| IFBench | 70.2% | — |
| Multilingual | ||
| MGSM | 83% | — |
| MMLU-ProX | 82.3% | — |
| NOVA-63 | 56.0% | — |
| INCLUDE | 83.3% | — |
| PolyMath | 43.1% | — |
| VWT2k-lite | 77.6% | — |
| MAXIFE | 72.8% | — |
| MathematicsSarvam 105B wins | ||
| AIME 2023 | 77% | — |
| AIME 2024 | 79% | — |
| AIME 2025 | 78% | 88.3% |
| AIME26 | 95.8% | — |
| AIME25 (Arcee) | 96.3% | — |
| HMMT Feb 2023 | 73% | — |
| HMMT Feb 2024 | 75% | — |
| HMMT Feb 2025 | 74% | — |
| HMMT Feb 2025 | 95.4% | 85.8% |
| HMMT Nov 2025 | 91.1% | 85.8% |
| HMMT Feb 2026 | 87.1% | — |
| MMAnswerBench | 81.8% | — |
| BRUMO 2025 | 76% | — |
| MATH-500 | 82% | 98.6% |
Kimi K2.5 is ahead overall, 72 to 60. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 76.8% and 45%.
Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 66.6. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Kimi K2.5 has the edge for coding in this comparison, averaging 66.7 versus 45. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Sarvam 105B has the edge for math in this comparison, averaging 92.3 versus 78.3. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
Kimi K2.5 has the edge for agentic tasks in this comparison, averaging 57.6 versus 49.5. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Kimi K2.5 has the edge for instruction following in this comparison, averaging 93.9 versus 84.8. Inside this category, IFEval 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.