---
format: "aidr-story-markdown/v1"
id: "128755116b8b9c05fe323feb2c16280847340b49674a612ecb437736e016082d"
canonical_url: "https://aidr.today/12875511?lang=en"
title: "Hugging Face Transformers v5.22.0"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-29T14:15:39.000Z"
category: "Models"
topics: ["llm"]
source_urls: []
summary: "Mistral Large 4 (Le Chonk) is Mistral's largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token, trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own datacenters in Europe. It unifies instruction following, reasoning, coding, and agentic workflows in a single set of weights and is natively fluent in 160+ languages, including every official language of the European Union."
---

# Hugging Face Transformers v5\.22\.0

> [Open the canonical story](<https://aidr.today/12875511?lang=en>)

**Published:** 2026-10-29T14:15:39.000Z
**Category:** Models
**Topics:** llm

## Summary

Mistral Large 4 \(Le Chonk\) is Mistral's largest and most capable model to date: a 1 trillion\-parameter natively multimodal Mixture\-of\-Experts model with 49 billion active parameters per token, trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own datacenters in Europe\. It unifies instruction following, reasoning, coding, and agentic workflows in a single set of weights and is natively fluent in 160\+ languages, including every official language of the European Union\.

## Sources

_No safe source URL was recorded._