Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
DeepSeek V3
49
1/8 categoriesSarvam 105B
60
Winner · 3/8 categoriesDeepSeek V3· Sarvam 105B
Pick Sarvam 105B if you want the stronger benchmark profile. DeepSeek V3 only becomes the better choice if instruction following is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
Sarvam 105B is clearly ahead on the aggregate, 60 to 49. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Sarvam 105B's sharpest advantage is in knowledge, where it averages 81.7 against 57.5. The single biggest benchmark swing on the page is MATH-500, 90.2% to 98.6%. DeepSeek V3 does hit back in instruction following, so the answer changes if that is the part of the workload you care about most.
DeepSeek V3 is also the more expensive model on tokens at $0.27 input / $1.10 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 DeepSeek V3 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 | DeepSeek V3 | Sarvam 105B |
|---|---|---|
| Agentic | ||
| BrowseComp | — | 49.5% |
| CodingSarvam 105B wins | ||
| LiveCodeBench | 37.6% | — |
| SWE-bench Verified | 42% | 45% |
| LiveCodeBench v6 | — | 71.7% |
| Multimodal & Grounded | ||
| Coming soon | ||
| Reasoning | ||
| LongBench v2 | 48.7% | — |
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| KnowledgeSarvam 105B wins | ||
| GPQA | 59.1% | — |
| MMLU-Pro | 75.9% | 81.7% |
| SimpleQA | 24.9% | — |
| MMLU | — | 90.6% |
| Instruction FollowingDeepSeek V3 wins | ||
| IFEval | 86.1% | 84.8% |
| Multilingual | ||
| Coming soon | ||
| MathematicsSarvam 105B wins | ||
| AIME 2024 | 39.2% | — |
| MATH-500 | 90.2% | 98.6% |
| AIME 2025 | — | 88.3% |
| HMMT Feb 2025 | — | 85.8% |
| HMMT Nov 2025 | — | 85.8% |
Sarvam 105B is ahead overall, 60 to 49. The biggest single separator in this matchup is MATH-500, where the scores are 90.2% and 98.6%.
Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 57.5. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
Sarvam 105B has the edge for coding in this comparison, averaging 45 versus 39.2. 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 90.2. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
DeepSeek V3 has the edge for instruction following in this comparison, averaging 86.1 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.