Back to Mistral models
Mistral logo

Mistral 7B Instruct v0.2

7B

by Mistral

A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length. An improved version of [Mistral 7B Instruct](/modelsmistralai/mistral-7b-instruct-v0.1), with the following changes: - 32k context window (vs 8k context in v0.1) - Rope-theta = 1e6 - No Sliding-Window Attention

Chat with Mistral 7B Instruct v0.2

Pricing

Input Tokens
Per 1M tokens
Free
Output Tokens
Per 1M tokens
Free
Image Processing
Per 1M tokens
$0.00/1M tokens

Supported Modalities

Input

text

Output

text

Performance Benchmarks

Intelligence Index
Overall intelligence score
1.0
GPQA
Graduate-level questions
17.7%
MMLU Pro
Multitask language understanding
24.5%
HLE
Human-like evaluation
4.3%
LiveCodeBench
Real-world coding tasks
4.6%
MATH 500
Mathematical problem solving
12.1%

Specifications

Context Length
33K tokens
Provider
Mistral
Throughput
120.047 tokens/s
Released
Dec 28, 2023
Model ID
mistralai/mistral-7b-instruct-v0.2

Ready to try it?

Start chatting with Mistral 7B Instruct v0.2 right now. No credit card required.

Start Chatting

More from Mistral

View all models