Mistral 7B Instruct v0.3

Permissive

Mistral AI · Mistral · Released 2024-05

7BText

The classic fully-permissive 7B baseline; not the strongest 7B anymore but has no usage restrictions at all.

Strengths

  • +True Apache 2.0, no gating or usage caps
  • +Function-calling support built in
  • +Fast on modest hardware

Limitations

  • -Outperformed by newer 7-8B models (Qwen2.5, Llama 3.1)
  • -32K context is smaller than most 2024+ peers

License

Apache 2.0

use commercially with attribution niceties

Hardware

runs on a good consumer GPU (or Apple Silicon) with quantization

Links

Stats

mistralai/Mistral-7B-Instruct-v0.3— downloads·— likes

via Hugging Face

Related models