Sarvam-105B
by Sarvam AI · family sarvam · listed Feb 2026RTW
Flagship Indian-language reasoning model for enterprise multilingual applications
Current rates USD per 1M tokens
Input—
Output—
Cache read—
Cache write—
Against the market
On the record
- Model id
sarvam-105b- Context window
- 131K tokens
- Max output
- 131K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- yes
- Knowledge cutoff
- —
- Released
- 2026-02-18
- Last updated
- 2026-03-06
Also listed at
- FastRouter$0.04 in · $0.16 out
- NanoGPT$0.045 in · $0.177 out