Sarvam-M 24B
By Sarvam AI · India
Updated 2026-07-13
Overview
Sarvam AI's 24B built on Mistral Small 3.1 with hybrid think/no-think modes, gaining +86% on romanized GSM-8K Indic and covering 11 Indian languages plus English.
When to pick this model
- Indic-language chat and content across 11 Indian languages
- Math-heavy workloads in romanized Indic scripts
- Hybrid reasoning where toggleable thinking helps
- Sovereign Indian deployments needing open weights
- Replacing closed APIs for Indian-market products
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 14 GB |
| Q5_K_M | 17 GB |
| Q8_0 | 26 GB |
| FP16 (no quantization) | 48 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, Sarvam-M 24B needs a 16 GB card at Q4_K_M (14 GB). Stepping up to Q8_0 nearly doubles the footprint to 26 GB, and unquantized FP16 weights take 48 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, Sarvam-M 24B needs roughly 24 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 4 tokens/sec on entry-level GPUs, on the order of 15 tokens/sec on a mid-range card, and up to 40 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches Sarvam-M 24B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for Sarvam-M 24B |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 14 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 14 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Q4_K_M (14 GB used) |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Q5_K_M (17 GB used) |
| 32 GB | RTX 5090 | Q8_0 (26 GB used) |
Which GPU should you buy to run Sarvam-M 24B?
To run Sarvam-M 24B locally at Q4, you need ~14 GB of VRAM. The best value for this is a RTX 5070 Ti (16 GB VRAM).
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Strengths
- +86% gain on romanized Indic GSM-8K
- Hybrid think/no-think mode toggle
- 11 Indian languages plus English
- Apache 2.0 with permissive commercial use
- Mistral Small 3.1 base brings solid general quality
Limitations
- No official Ollama distribution yet
- Strong Indic focus limits broader multilingual use
- Smaller community ecosystem than Mistral mainline
Typical workloads
In our catalog grid, Sarvam-M 24B is filed under Indian Languages, Indic Reasoning — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multi-step reasoning and math-flavoured tasks; multilingual workloads.
The 32k-token context window covers long chats and mid-sized documents, though very large retrieval workloads will need chunking. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.
Architecture & training
Architecture: Dense 24B · Mistral Small 3.1 base · hybrid think/non-think
Training: 11 Indian languages + EN.
The top open model for Indic markets — pick it when you need real Indian-language coverage with hybrid reasoning.
Quick start
# HuggingFace : sarvamai/sarvam-mOr use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Similar models worth comparing
Frequently asked questions
How much VRAM does Sarvam-M 24B need?
At the recommended Q4_K_M quantization, Sarvam-M 24B needs about 14 GB of VRAM. Q8_0 takes 26 GB, and unquantized FP16 weights take 48 GB.
Can Sarvam-M 24B run without a GPU?
Yes — with roughly 24 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does Sarvam-M 24B support?
Sarvam-M 24B supports a 32k-token context window (32,768 tokens).
Can I use Sarvam-M 24B commercially?
Yes. Sarvam-M 24B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.
How fast is Sarvam-M 24B on consumer hardware?
Our compatibility engine estimates on the order of 15 tokens/sec on a mid-range GPU and up to 40 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of Sarvam-M 24B should I download first?
Start with Q4_K_M (14 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. On a 24 GB card you can run up to Q5_K_M.