mistral

The Mistral 7B model released by Mistral AI

68.7K Pulls Updated 6 weeks ago

Mistral 7B model is an Apache licensed 7.3B parameter model. It is available in both instruct (instruction following) and text completion.

The Mistral AI team has noted that Mistral 7B:

  • Outperforms Llama 2 13B on all benchmarks
  • Outperforms Llama 1 34B on many benchmarks
  • Approaches CodeLlama 7B performance on code, while remaining good at English tasks
  • Uses Grouped-query attention (GQA) for faster inference
  • Uses Sliding Window Attention (SWA) to handle longer sequences at smaller cost

For benchmark data, please visit the Mistral AI news release under References.

CLI

Instruct:

ollama run mistral

Text completion:

ollama run mistral:text

API

Example:

curl -X POST http://localhost:11434/api/generate -d '{
  "model": "mistral",
  "prompt":"Here is a story about llamas eating grass"
 }'

Variations

instruct Instruct models follow instructions
text Text models are the base foundation model without any fine-tuning for conversations, and are best used for simple text completion.

Memory requirements

  • 7b models generally require at least 8GB of RAM

References

HuggingFace

Mistral AI News Release