What are the hardware requirements for MetaMath Mistral 7B?

MetaMath Mistral 7B requires approximately 14.4 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 MetaMath Mistral 7B and how large is it?

MetaMath Mistral 7B is developed by meta-math, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does MetaMath Mistral 7B perform on standard benchmarks?

MetaMath Mistral 7B has the following published scores: MMLU 61.95. Compare against reference models on this page or on the LLM Explorer leaderboards.

MetaMath Mistral 7B by meta-math — VRAM 14.4GB, 32K context

Name: MetaMath Mistral 7B
Author: meta-math

MetaMath Mistral 7B is an open-source language model by meta-math. Features: 7b LLM, VRAM: 14.4GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.12, Arc: 60.7, HellaSwag: 82.6, MMLU: 62, GSM8K: 68.8.

Arxiv:2309.12284 Arxiv:2310.06825 Dataset:meta-math/metamathqa Endpoints compatible Mistral Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/meta-math/MetaMath-Mistral-7B

MetaMath Mistral 7B Parameters and Internals

Model Type

text generation

Use Cases

Areas:

mathematics, AI research

Training Details

Data Sources:

meta-math/MetaMathQA, GSM8K, MATH

Methodology:

Fine-tuning with adjusted learning rate

Model Architecture:

Mistral-7B

Input Output

Input Format:

{instruction}

Accepted Modalities:

text

Output Format:

Generated text response

Performance Tips:

Using a smaller learning rate when fine-tuning

LLM Name	MetaMath Mistral 7B
Repository 🤗	https://huggingface.co/meta-math/MetaMath-Mistral-7B
Model Size	7b
Required VRAM	14.4 GB
Updated	2026-07-31
Maintainer	meta-math
Model Type	mistral
Model Files	9.9 GB: 1-of-2 4.5 GB: 2-of-2
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.35.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	[PAD]
Vocabulary Size	32001
Torch Data Type	bfloat16

Rank the MetaMath Mistral 7B capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Quantized Models of the MetaMath Mistral 7B

Model	Likes	Downloads	VRAM
MetaMath Mistral 7B GGUF	10	752	3 GB
MetaMath Mistral 7B GPTQ	3	10	4 GB
MetaMath Mistral 7B AWQ	1	13	4 GB

Best Alternatives to MetaMath Mistral 7B

Best Alternatives	Context / RAM	Downloads	Likes
...Nemo Instruct 2407 Abliterated	1000K / 24.5 GB	220	20
MegaBeam Mistral 7B 512K	512K / 14.4 GB	8470	54
SpydazWeb AI HumanAI RP	512K / 14.4 GB	16	1
SpydazWeb AI HumanAI 002	512K / 14.4 GB	18	1
...daz Web AI ChatML 512K Project	512K / 14.5 GB	12	0
MegaBeam Mistral 7B 300K	282K / 14.4 GB	3779	16
MegaBeam Mistral 7B 300K	282K / 14.4 GB	8018	17
Hebrew Mistral 7B 200K	256K / 30 GB	23	15
Astral 256K 7B	250K / 14.4 GB	9	0
Astral 256K 7B V2	250K / 14.4 GB	8	0

Note: green Score (e.g. "73.2") means that the model is better than meta-math/MetaMath-Mistral-7B.