Sarvam AI · India · Released 2025-05

Sarvam-M

India's flagship 24B open model with strong Indic-language reasoning and math performance.

Official site

Access

open-weights

Context

32,000 tokens

Modalities

text

Languages

10 Indic + English

Benchmark scores

MMLU

74.7%

Massive Multitask Language Understanding

GPQA

N/A

Graduate-level science QA (Diamond)

HumanEval

N/A

Python code completion

SWE-bench

N/A

Real GitHub issue resolution (Verified)

LiveCodeBench

N/A

Contamination-free coding

AIME

N/A

American Invitational Mathematics Exam

MMMU

N/A

Multimodal college-level reasoning

MATH

71.4%

Competition-level math word problems

Scores as reported by the vendor or leading public leaderboards. "N/A" means the score has not been publicly disclosed for this metric.

Explain like I'm 5

An Indian model that understands Hindi, Tamil, Bengali and more really well.

Key features

10 Indian languages
Hybrid reasoning mode
Open weights

Strengths

Indic languages
Open weights

Limitations

Smaller general benchmarks vs frontier

Best for

Bharat-focused appsIndic RAG

Related models

OpenAI

GPT-5

OpenAI's flagship reasoning + multimodal model with a unified router that switches between fast and deep-thinking modes.

Anthropic

Claude Sonnet 4.5

Anthropic's best coding + agentic model, purpose-built for long-horizon computer-use and software engineering tasks.

Google DeepMind

Gemini 2.5 Pro

Google's most capable model with native 1M-token context and full multimodal I/O across text, images, audio and video.

xAI

Grok 4

xAI's frontier reasoning model with real-time X data grounding and a heavy multi-agent variant (Grok 4 Heavy).