Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
DeepSeek-R1
45
1/8 categoriesSarvam 30B
48
Winner · 3/8 categoriesDeepSeek-R1· Sarvam 30B
Pick Sarvam 30B if you want the stronger benchmark profile. DeepSeek-R1 only becomes the better choice if agentic is the priority or you need the larger 128K context window.
Sarvam 30B has the cleaner overall profile here, landing at 48 versus 45. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
Sarvam 30B's sharpest advantage is in knowledge, where it averages 80 against 47. The single biggest benchmark swing on the page is AIME 2025, 45% to 80%. DeepSeek-R1 does hit back in agentic, so the answer changes if that is the part of the workload you care about most.
DeepSeek-R1 is also the more expensive model on tokens at $0.55 input / $2.19 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Sarvam 30B. That is roughly Infinityx on output cost alone. DeepSeek-R1 gives you the larger context window at 128K, compared with 64K for Sarvam 30B.
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-R1 | Sarvam 30B |
|---|---|---|
| AgenticDeepSeek-R1 wins | ||
| Terminal-Bench 2.0 | 42% | — |
| BrowseComp | 49% | 35.5% |
| OSWorld-Verified | 44% | — |
| CodingSarvam 30B wins | ||
| HumanEval | 92% | 92.1% |
| SWE-bench Verified | 49.2% | 34% |
| LiveCodeBench | 19% | — |
| SWE-bench Pro | 25% | — |
| LiveCodeBench v6 | — | 70.0% |
| Multimodal & Grounded | ||
| MMMU-Pro | 43% | — |
| OfficeQA Pro | 53% | — |
| Reasoning | ||
| MuSR | 40% | — |
| BBH | 66% | — |
| LongBench v2 | 58% | — |
| MRCRv2 | 57% | — |
| ARC-AGI-2 | 1.3% | — |
| gpqaDiamond | — | 66.5% |
| KnowledgeSarvam 30B wins | ||
| MMLU | 90.8% | 85.1% |
| GPQA | 71.5% | — |
| SuperGPQA | 41% | — |
| MMLU-Pro | 84% | 80% |
| HLE | 14% | — |
| FrontierScience | 44% | — |
| SimpleQA | 30.1% | — |
| Instruction Following | ||
| IFEval | 83.3% | — |
| Multilingual | ||
| MGSM | 61% | — |
| MMLU-ProX | 60% | — |
| MathematicsSarvam 30B wins | ||
| AIME 2023 | 44% | — |
| AIME 2024 | 79.8% | — |
| AIME 2025 | 45% | 80% |
| HMMT Feb 2023 | 40% | — |
| HMMT Feb 2024 | 42% | — |
| HMMT Feb 2025 | 41% | — |
| BRUMO 2025 | 43% | — |
| MATH-500 | 97.3% | 97% |
| HMMT Feb 2025 | — | 73.3% |
| HMMT Nov 2025 | — | 74.2% |
Sarvam 30B is ahead overall, 48 to 45. The biggest single separator in this matchup is AIME 2025, where the scores are 45% and 80%.
Sarvam 30B has the edge for knowledge tasks in this comparison, averaging 80 versus 47. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Sarvam 30B has the edge for coding in this comparison, averaging 34 versus 28.3. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
Sarvam 30B has the edge for math in this comparison, averaging 86.5 versus 57.4. Inside this category, AIME 2025 is the benchmark that creates the most daylight between them.
DeepSeek-R1 has the edge for agentic tasks in this comparison, averaging 44.5 versus 35.5. Inside this category, BrowseComp 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.