What are the hardware requirements for Sarvam M?

Sarvam M requires approximately 47.3 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Sarvam M and how large is it?

Sarvam M is developed by sarvamai, a model with 24b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Sarvam M?

Sarvam M is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Sarvam M by sarvamai — VRAM 47.3GB, 32K context

Name: Sarvam M
Author: sarvamai

Sarvam M is an open-source language model by sarvamai. Features: 24b LLM, VRAM: 47.3GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.25.

Base model:finetune:mistralai/... Base model:mistralai/mistral-s... Bn Conversational En Endpoints compatible Gu Hi Kn Mistral Ml Mr Or Pa Region:us Safetensors Sharded Ta Te Tensorflow

Model Card on HF 🤗: https://huggingface.co/sarvamai/sarvam-m

Sarvam M Benchmarks

LLME Score: 0.25487

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

🌟 Advertise your project 🚀

Sarvam M Parameters and Internals

LLM Name	Sarvam M
Repository 🤗	https://huggingface.co/sarvamai/sarvam-m
Base Model(s)	mistralai/Mistral-Small-3.1-24B-Base-2503 mistralai/Mistral-Small-3.1-24B-Base-2503
Model Size	24b
Required VRAM	47.3 GB
Updated	2026-05-05
Maintainer	sarvamai
Model Type	mistral
Model Files	4.8 GB: 1-of-10 4.8 GB: 2-of-10 4.8 GB: 3-of-10 4.9 GB: 4-of-10 4.8 GB: 5-of-10 4.8 GB: 6-of-10 4.9 GB: 7-of-10 4.8 GB: 8-of-10 4.8 GB: 9-of-10 3.9 GB: 10-of-10
Supported Languages	en bn hi kn gu mr ml or pa ta te
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.51.3
Tokenizer Class	LlamaTokenizerFast
Vocabulary Size	131072
Torch Data Type	bfloat16

Quantized Models of the Sarvam M

Model	Likes	Downloads	VRAM
Sarvam M 24B Q5 K M Gguf	2	69	16 GB

Best Alternatives to Sarvam M

Best Alternatives	Context / RAM	Downloads	Likes
...ral Nemo Instruct 24B Merge V1	1000K / 24.6 GB	9	0
Cydonia 24B V4.3	128K / 47.3 GB	3321	87
Sketch Cydonia 24B V1.3	128K / 47.3 GB	118	4
Cogidonia 24B	128K / 95.3 GB	149	1
Sketch Cydonia	128K / 47.3 GB	143	9
Sketch Cydonia 24B V1.2	128K / 47.3 GB	68	8
Magistaroth 24B V1.1	128K / 47.3 GB	84	5
Magidonia 24B V4.3	128K / 47.3 GB	265	60
Cydonia 24B V4.3 Heretic V4	128K / 47.3 GB	111	5
...donia 24B V4.3 Absolute Heresy	128K / 47.1 GB	141	14

Note: green Score (e.g. "73.2") means that the model is better than sarvamai/sarvam-m.

Rank the Sarvam M Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 53518 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Sarvam M by sarvamai