Back to Mixtral
Mistral AI's landmark 8x7B MoE instruct model — sparse mixture-of-experts with GPT-3.5-level performance at fraction of the compute.
33K tokensFree / Open weightsMoEApache 2.0
Benchmarks
Mixtral-8x7B-Instruct-v0.1
20.6%Mixtral-8x22B-v0.1
7.3%🧟🧟 (Mistral-Small-3.1-24B) — close family
