Mistral 7B Instruct v0.1

Open Source

Mistral AI

A 7.3B parameter model that outperforms Llama 2 13B on all benchmarks, with optimizations for speed and context length.

openrouter.ai ↗

📖Context4K tokens

🚀Speed-

💵Input price$0.110/1M

💸Output price$0.190/1M

🧠Parameters7.242B

Benchmarks

Gray bar = dataset average

Composite ScoreScore = MMLU×20% + GPQA×30% + HumanEval×25% + SWE-Bench×25%-

HF Open LLM Leaderboard

IFEval · BBH · MATH · GPQA · MuSR · MMLU-Pro

Average

12.8

View on HF →

IFEvalInstruction following

44.9%avg 79.9

BBHBig Bench Hard

7.7%avg 48.7

MATH Lvl 5Competition mathematics

2.3%avg 37.8

GPQAGraduate-level science Q&A

0.0%avg 12.9

MuSRMultistep soft reasoning

6.1%avg 10.7

MMLU-ProProfessional knowledge

15.7%avg 40.0

Made by

More from Mistral AI

Mistral Small 3.2 24B

API Providers

Details

Provider: Mistral AI
Parameters: 7.242B
Context: 4K tokens
Speed: -
Open Source: Yes
License: Open (varies)
Released: Sept 2023

Reviews

No reviews yet

← View all 376 models