MODELS

mistralai/Mistral-7B-Instruct-v0.3

by Mistral AI

modellead-sourcehugging-face-popular-modelssource:github.commistral-aimistral-7binstructopen-modelhugging-facegithub

Overview

Open 7B Mistral AI instruct model with 32k context, listed on the official Mistral 7B v0.3 model card and hosted on Hugging Face.

Details

Mistral AI’s documentation identifies Mistral 7B v0.3 as an open 7B model with a 32k context window and links directly to the Mistral-7B-Instruct-v0.3 weights on Hugging Face. The official mistral-inference GitHub repository is described as the official inference library for Mistral models and provides installation instructions plus a direct download link for the 7B Instruct v0.3 model tarball with an md5 checksum.

When to Use

Use when you need an open 7B Mistral instruct model referenced by Mistral AI’s official model card. Use when you want to run or evaluate Mistral 7B Instruct v0.3 through Mistral AI’s official mistral-inference library. Use when a 32k-context 7B model is appropriate for prototyping comparison or local inference experiments.

Getting Started

Read the official Mistral 7B v0.3 model card in Mistral AI docs.
Open the Hugging Face page for mistralai/Mistral-7B-Instruct-v0.3 to access the Instruct weights.
Follow the installation instructions in the official mistral-inference GitHub repository.
If using the repository download flow
use the provided 7B Instruct v0.3 tarball link and md5 checksum from the official repository metadata.

Key Features

•Open 7B model according to Mistral AI’s official model card.
•32k context window listed in the official Mistral 7B v0.3 documentation.
•Instruct weights available on Hugging Face at mistralai/Mistral-7B-Instruct-v0.3.
•Supported by Mistral AI’s official mistral-inference library.

Capabilities

•instruction-following text generation
•32k-context language model evaluation
•local inference via official Mistral inference library
•Hugging Face model weight access

Last updated Jun 7, 2026