Mixtral 8x7B

Active

Open-weights mixture-of-experts model with GPT-3.5 class performance.

Overview

Mixtral 8x7B uses a sparse mixture-of-experts architecture (8 experts, 2 active per token) to deliver GPT-3.5 class performance at lower inference cost than a dense 45B model. Apache-2.0 licensed.

Integrations & tooling support

Tool calling: Supported
Structured outputs: Not supported

Price vs quality

Not enough data

This model has no benchmark scores recorded yet.

Community ratings

No ratings yet. Be the first to rate Mixtral 8x7B.