All tasks / Models
Sarvam-105B
Sarvam AI's open-weights MoE (Apache-2.0) focused on Indian languages, 105B total and 10.3B active.
- Sarvam AI
- Open weights
- Released 3 Mar 2026
Where it ranks
- CChat & assistantsModels · Local · 9 on Artificial Analysis Intelligence Index (secondary benchmark)
Facts
- Vendor
- Sarvam AI
- Family
- Sarvam
- Released
- 3 Mar 2026
- Weights
- Open: download and run it yourself
- License
- Apache-2.0
- Parameters
- 106B (10.3B active)
- GPU memory to run locally
- ~62 GB (4-bit ≈ 106 B × 0.57 GB + 1.5 GB ≈ 62 GB, estimate.)
- Access
- Free (open weights)
- Last verified
- 19 Oct 2026
Price from huggingface.co, 19 Oct 2026.
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Intelligence Index | 9 | Sarvam 105B (High)Raw value from the page's embedded model data; the site displays it rounded to an integer. Flagged 'estimated' by Artificial Analysis. | 19 Oct 2026 | artificialanalysis.ai |
Where you can use it
No tracked tool lists this model yet; as open weights it runs in local runtimes.