AI PLATFORMS & MODELS
OPEN WEIGHTSOPEN SOURCE SELF-HOSTABLE

Mixtral 8x22B

by Mistral AI

The open MoE that started the wave.

80HYLAND SCORE
2.9k views this week
Try Mixtral 8x22B
SHARE Post on X LinkedIn

Overview

Mixtral 8x22B is Mistral’s open-weight sparse mixture-of-experts model under Apache 2.0. A proven workhorse for self-hosted chat, RAG, and fine-tuning with 39B active parameters.

Key features

  • 141B MoE with 39B active
  • Apache 2.0 license
  • 64K context window
  • Mature fine-tuning ecosystem
  • Multilingual (EN/FR/DE/ES/IT)

Tags

Self-HostMoEApache 2.0Fine-tuning

Related in AI Platforms & Models

FEATURED VERIFIED

DeepSeek R1

DeepSeek
93HYLAND

Open-weight reasoning that rivals the frontier.

Open WeightsOpen Source Self-host
15.3k/wk
View
VERIFIED
92HYLAND

Self-hosted AI, privacy-critical apps

Open WeightsOpen Source Self-host
View
FEATURED VERIFIED

Kimi K2

Moonshot AI
91HYLAND

Open trillion-parameter agentic intelligence.

Open WeightsOpen Source Self-host
12.8k/wk
View