IndiVibe
Sarvam 105B 90% Fading
Models answer 110 source prompts in chat, STEM, math and coding, translated into 22 Indian languages in native and romanized script. A judge model compares answers pairwise.
The judge scores fluency, language and script correctness, usefulness and verbosity. Claude Opus 4 reference answers in English guide the usefulness score. Results are win rates.
Made by Sarvam AI Released Mar 2026 Chat and instructions in many languages All tasks public Last in a launch post Mar 2026 , dropped after it saturated
Last model added Mar 2026
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
2
Models scored in lab cards
2
Launch posts list it, from 1 lab
Sarvam 105B leads with 90%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 Sarvam 105B 90%
- 2 Sarvam 30B 89%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Sarvam 105B | 90% | Mar 2026 | Model card | |
| 2 | Sarvam 30B | 89% | Mar 2026 | Model card |
The best score passed 85% in Mar 2026
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Reported by the lab in its model card
Labs headline it less often now, last in Mar 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 6 Mar 2026 | Sarvam 30B | 89 | Average pairwise win rate across all dimensions on the Indic benchmark (indivibe) |
| 6 Mar 2026 | Sarvam 105B | 90 | Average pairwise win rate across all dimensions on the Indic benchmark (indivibe) |
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5% - MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87% - CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7% - Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy.
Inkling 88.7% - MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%