# Hugging Face — Welcome Mixtral - a SOTA Mixture of Experts on Hugging Face

- Company: Hugging Face (huggingface.co)
- Announced: 2023-12-11
- Category: new-model
- Coverage: not counted
- Announcement: yes
- Group: models
- Source: https://huggingface.co/blog/mixtral
- Record: https://forck.live/items/1966-welcome-mixtral-a-sota-mixture-of-experts-on-hugging-face
- Subject: Platform
- Open weights: yes
- Models affected: Mixtral 8x7b, Mixtral-8x7B-v0.1, Mixtral-8x7B-Instruct-v0.1
- License: Apache 2.0
- Context window: 32k tokens

Mistral released Mixtral 8x7b, a Mixture-of-Experts open-access model that outperforms GPT-3.5 and Llama 2 70B on many benchmarks, with Apache 2.0 license and 32k context length. Hugging Face announced comprehensive integration including models on the Hub, Transformers, Inference Endpoints, and Text Generation Inference.

## Evidence

Verbatim from https://huggingface.co/blog/mixtral:

> Mixtral 8x7b is an exciting large language model released by Mistral today, which sets a new state-of-the-art for open-access models and outperforms GPT-3.5 across many benchmarks.

---

Record: https://forck.live/items/1966-welcome-mixtral-a-sota-mixture-of-experts-on-hugging-face
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md