Sarvam-M places in the top three on TriviaQA.
Open weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TriviaQA | 91% | #2 of 40 | Lab model card | |
| ARC-Challenge | 95% | #8 of 49 | Lab model card | |
| MMLU | 87% | #42 of 131 | Lab model card | |
| IFEval | 87% | #63 of 135 | Lab model card | |
| MT-Bench | 8.14 | #21 of 23 | Lab model card | |
| IndicGenBench | 37.2 | #12 of 13 | Official leaderboard | |
| Global PIQA | 0.13% | #44 of 44 | Official leaderboard | |
| MILU | 75% | #11 of 11 | Lab model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 94% | #20 of 92 | Lab model card | |
| MATH | 81% | #25 of 99 | Lab model card | |
| GPQA Diamond | 48% | #252 of 302 | Lab model card |
The Self-Reported Index fits one score to the 10 capability results Sarvam AI reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 124.0 to 130.1. How we compute it.
The benchmarks the lab chose to headline, with the scores it gave.