Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Grok 4
68
Winner · 2/8 categoriesSarvam 105B
60
3/8 categoriesGrok 4· Sarvam 105B
Pick Grok 4 if you want the stronger benchmark profile. Sarvam 105B only becomes the better choice if knowledge is the priority or you want the stronger reasoning-first profile.
Grok 4 is clearly ahead on the aggregate, 68 to 60. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Grok 4's sharpest advantage is in coding, where it averages 65.8 against 45. The single biggest benchmark swing on the page is SWE-bench Verified, 73% 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.
Sarvam 105B is the reasoning model in the pair, while Grok 4 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 | Grok 4 | Sarvam 105B |
|---|---|---|
| AgenticGrok 4 wins | ||
| Terminal-Bench 2.0 | 57% | — |
| BrowseComp | 63% | 49.5% |
| OSWorld-Verified | 56% | — |
| CodingGrok 4 wins | ||
| HumanEval | 79% | — |
| SWE-bench Verified | 73% | 45% |
| LiveCodeBench | 79.4% | — |
| SWE-bench Pro | 48% | — |
| React Native Evals | 70.1% | — |
| LiveCodeBench v6 | — | 71.7% |
| Multimodal & Grounded | ||
| MMMU-Pro | 80% | — |
| OfficeQA Pro | 76% | — |
| Reasoning | ||
| MuSR | 81% | — |
| BBH | 83% | — |
| LongBench v2 | 72% | — |
| MRCRv2 | 71% | — |
| ARC-AGI-2 | 16% | — |
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| KnowledgeSarvam 105B wins | ||
| MMLU | 87% | 90.6% |
| GPQA | 86% | — |
| SuperGPQA | 84% | — |
| MMLU-Pro | 77% | 81.7% |
| HLE | 16% | — |
| FrontierScience | 75% | — |
| SimpleQA | 83% | — |
| Instruction FollowingSarvam 105B wins | ||
| IFEval | 82% | 84.8% |
| Multilingual | ||
| MGSM | 84% | — |
| MMLU-ProX | 79% | — |
| MathematicsSarvam 105B wins | ||
| AIME 2023 | 87% | — |
| AIME 2024 | 89% | — |
| AIME 2025 | 88% | 88.3% |
| HMMT Feb 2023 | 84% | — |
| HMMT Feb 2024 | 86% | — |
| HMMT Feb 2025 | 85% | — |
| BRUMO 2025 | 87% | — |
| MATH-500 | 83% | 98.6% |
| HMMT Feb 2025 | — | 85.8% |
| HMMT Nov 2025 | — | 85.8% |
Grok 4 is ahead overall, 68 to 60. The biggest single separator in this matchup is SWE-bench Verified, where the scores are 73% and 45%.
Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 65.3. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
Grok 4 has the edge for coding in this comparison, averaging 65.8 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 86.4. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
Grok 4 has the edge for agentic tasks in this comparison, averaging 58.1 versus 49.5. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Sarvam 105B has the edge for instruction following in this comparison, averaging 84.8 versus 82. 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.