SA
CompareSarvam 30B
sarvam/sarvam-30b
Efficient Indian-language reasoning model for chat, coding, and multilingual work
- Context window
- 131K
- Input
- Text
- Output
- Text
- Released
- 2026-02-18
- Max output
- 128K
- Model parameters
- 32B
- Weights
- Open
- License
- apache-2.0
- Capabilities
- Open weight, Reasoning, Temperature, Tool call
Pricing
No official pricing available.
Performance
Overall capability
36.4/100
- Coding
- —
- Reasoning
- 37.0
- Professional work
- —
- Knowledge & language
- 14.0
| IFBench（Artificial Analysis）Sarvam 30B (High) | Knowledge & language | Artificial Analysis | 2026-11-05 | |
| Humanity’s Last ExamSarvam 30B (High) | Reasoning | Artificial Analysis | 2026-11-05 | |
| AA-LCRSarvam 30B (High) | Knowledge & language | Artificial Analysis | 2026-11-05 | |
| CritPtSarvam 30B (High) | Reasoning | Artificial Analysis | 2026-11-05 | |
| AA-OmniscienceSarvam 30B (High) | Knowledge & language | Artificial Analysis | 2026-10-27 |
Providers
| Cache write | |||||||
|---|---|---|---|---|---|---|---|
| FastRouter | sarvam/sarvam-30b | Openai chat | - | - | - | ||
| NanoGPT | sarvam-30b | Openai chat | - | - | |||
| Sarvam AI | sarvam-30b | Custom | - | - | - | - | - |
ModelRegistry · Catalog updated 2026-11-05