Sarvam 30B

Released
2026-03-22
Last refreshed
2026-09-08
Status
Researched 134d ago
Open sourceCommercial use: permittedAgentsJSON / Tool use

Sarvam 30B is a released agents and json / tool use model with open-source; evaluate it while provider pricing coverage matures.

Sarvam.ai releases · 1 in the last 12 monthsChangelog →
Specifications
Family
Sarvam
Released
2026-03-22
Context
66k
Parameters
30B (2.4B active)
Architecture
Mixture of Experts
Knowledge cutoff
2025-06
Specialization
reasoning
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Available
Code
Unknown
Training
Pretrained
Created by

Multilingual Voice AI for India

Bengaluru, India
Founded 2023
Website
Pricing

No tracked provider token pricing is available yet.

About

Sarvam 30B is an open-source MoE reasoning model with 30B total parameters. Optimized for math, coding, and knowledge tasks with state-of-the-art performance across 22 Indian languages. Supports multilingual voice calls and streaming inference.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

JSON / Tool use

Benchmark peer barsfor Agents

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

Compare Sarvam 30B with other models