Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Qwen3 235B 2507 (Reasoning)
55
Winner · 2/8 categoriesSarvam 30B
48
2/8 categoriesQwen3 235B 2507 (Reasoning)· Sarvam 30B
Pick Qwen3 235B 2507 (Reasoning) if you want the stronger benchmark profile. Sarvam 30B only becomes the better choice if knowledge is the priority.
Qwen3 235B 2507 (Reasoning) is clearly ahead on the aggregate, 55 to 48. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3 235B 2507 (Reasoning)'s sharpest advantage is in agentic, where it averages 47.4 against 35.5. The single biggest benchmark swing on the page is HumanEval, 32% to 92.1%. Sarvam 30B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Qwen3 235B 2507 (Reasoning) 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 | Qwen3 235B 2507 (Reasoning) | Sarvam 30B |
|---|---|---|
| AgenticQwen3 235B 2507 (Reasoning) wins | ||
| Terminal-Bench 2.0 | 47% | — |
| BrowseComp | 48% | 35.5% |
| CodingQwen3 235B 2507 (Reasoning) wins | ||
| HumanEval | 32% | 92.1% |
| SWE-bench Verified | 16% | 34% |
| LiveCodeBench | 74.1% | — |
| SWE-bench Pro | 29% | — |
| LiveCodeBench v6 | — | 70.0% |
| Multimodal & Grounded | ||
| MMMU-Pro | 38% | — |
| OfficeQA Pro | 47% | — |
| Reasoning | ||
| MuSR | 36% | — |
| BBH | 63% | — |
| LongBench v2 | 58% | — |
| MRCRv2 | 58% | — |
| gpqaDiamond | — | 66.5% |
| KnowledgeSarvam 30B wins | ||
| MMLU | 40% | 85.1% |
| GPQA | 81.1% | — |
| SuperGPQA | 64.9% | — |
| MMLU-Pro | 84.4% | 80% |
| HLE | 6% | — |
| FrontierScience | 42% | — |
| SimpleQA | 38% | — |
| Instruction Following | ||
| IFEval | 87.8% | — |
| Multilingual | ||
| MGSM | 62% | — |
| MMLU-ProX | 81% | — |
| MathematicsSarvam 30B wins | ||
| AIME 2023 | 40% | — |
| AIME 2024 | 42% | — |
| AIME 2025 | 92.3% | 80% |
| HMMT Feb 2023 | 36% | — |
| HMMT Feb 2024 | 38% | — |
| HMMT Feb 2025 | 37% | — |
| BRUMO 2025 | 39% | — |
| MATH-500 | 60% | 97% |
| HMMT Feb 2025 | — | 73.3% |
| HMMT Nov 2025 | — | 74.2% |
Qwen3 235B 2507 (Reasoning) is ahead overall, 55 to 48. The biggest single separator in this matchup is HumanEval, where the scores are 32% and 92.1%.
Sarvam 30B has the edge for knowledge tasks in this comparison, averaging 80 versus 50. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 (Reasoning) has the edge for coding in this comparison, averaging 43.3 versus 34. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Sarvam 30B has the edge for math in this comparison, averaging 86.5 versus 65.6. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 (Reasoning) has the edge for agentic tasks in this comparison, averaging 47.4 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.