LLM
GA
Open weights
Sarvam 105B
Sovereign open model for Indian languages
Release2026-01
Context131K
Max output—
Parameters105B
Local / VRAM120–160 GB
LicenseOpen weights
Overview
Sarvam 105B is a model in the LLM category from Sarvam AI. It was released in 2026-01. The vendor is based in India. Best for: Sovereign open model for Indian languages.
Strengths
- Indian languages
- Apache-2.0
- MoE
Limits
- No vision
- Heavy hardware
Specifications
| Vendor | Sarvam AI |
|---|---|
| Category | LLM |
| Release | 2026-01 |
| Status | GA |
| License | Open weights |
| License name | Apache-2.0 |
| Commercial use | Allowed |
| Parameters | 105B |
| Active parameters | 10B |
| Architecture | MoE transformer |
| Context window | 131K |
| Input modalities | Text |
| Output modalities | Text |
| Prompt types | Text |
| Negative prompts | No |
| Max outputs | 1 |
| Local deployment | Yes |
| VRAM | 120–160 GB |
| System RAM | 160–320 GB |
| API access | Yes |
| Pricing | Free weights |
| Confidence | Medium |
| Vendor country | India |
| Vendor type | Open lab |
Pricing & hardware
Free weights
This model can be self-hosted or run locally. Approximate GPU memory: 120–160 GB. Approximate system RAM: 160–320 GB. API access is available.