Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
GPT-4.1 mini
57
2/8 categoriesSarvam 105B
60
Winner · 2/8 categoriesGPT-4.1 mini· Sarvam 105B
Pick Sarvam 105B if you want the stronger benchmark profile. GPT-4.1 mini only becomes the better choice if agentic is the priority or you need the larger 1M context window.
Sarvam 105B has the cleaner overall profile here, landing at 60 versus 57. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
Sarvam 105B's sharpest advantage is in knowledge, where it averages 81.7 against 64.2. The single biggest benchmark swing on the page is BrowseComp, 71% to 49.5%. GPT-4.1 mini does hit back in agentic, so the answer changes if that is the part of the workload you care about most.
GPT-4.1 mini is also the more expensive model on tokens at $0.40 input / $1.60 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 GPT-4.1 mini 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. GPT-4.1 mini gives you the larger context window at 1M, compared with 128K for Sarvam 105B.
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 | GPT-4.1 mini | Sarvam 105B |
|---|---|---|
| AgenticGPT-4.1 mini wins | ||
| Terminal-Bench 2.0 | 54% | — |
| BrowseComp | 71% | 49.5% |
| OSWorld-Verified | 49% | — |
| CodingSarvam 105B wins | ||
| SWE-bench Verified | 23.6% | 45% |
| SWE-bench Pro | 30% | — |
| LiveCodeBench v6 | — | 71.7% |
| Multimodal & Grounded | ||
| OfficeQA Pro | 74% | — |
| Reasoning | ||
| LongBench v2 | 80% | — |
| MRCRv2 | 82% | — |
| gpqaDiamond | — | 78.7% |
| hle | — | 11.2% |
| KnowledgeSarvam 105B wins | ||
| MMLU | 87.5% | 90.6% |
| GPQA | 64.2% | — |
| MMLU-Pro | — | 81.7% |
| Instruction FollowingGPT-4.1 mini wins | ||
| IFEval | 88.5% | 84.8% |
| Multilingual | ||
| MMLU-ProX | 72% | — |
| Mathematics | ||
| AIME 2024 | 23.1% | — |
| MATH-500 | — | 98.6% |
| AIME 2025 | — | 88.3% |
| HMMT Feb 2025 | — | 85.8% |
| HMMT Nov 2025 | — | 85.8% |
Sarvam 105B is ahead overall, 60 to 57. The biggest single separator in this matchup is BrowseComp, where the scores are 71% and 49.5%.
Sarvam 105B has the edge for knowledge tasks in this comparison, averaging 81.7 versus 64.2. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Sarvam 105B has the edge for coding in this comparison, averaging 45 versus 27.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.
GPT-4.1 mini has the edge for agentic tasks in this comparison, averaging 56.5 versus 49.5. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
GPT-4.1 mini has the edge for instruction following in this comparison, averaging 88.5 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.