How to run Mistral Large 4 on your own machine
How to run Mistral Large 4 on your own machine. The open weights of Mistral Large 4 are available for download on Hugging Face as of October 29, 2026. You get the 1.05T parameter model that Mistral serves through their API under a modified MIT license, in both FP8 and NVFP4 checkpoint formats with GGUF quants provided by the community available for download without any approval process. It can be served on 8 datacenter GPUs with vLLM or SGLang…
Mistral Large 4, a 1.05 trillion parameter AI model, has been made available for download on Hugging Face. The model, also known as "Le Chonk", is a multimodal mixture-of-experts with 49B parameters per token and a 1M token context window. It was released under the Mistral Ecosystem Open-Weight License on October 29, 2024. Users can run the model in three ways: by calling the Mistral API for $0.70 per million input tokens, using a hosted provider, or by downloading and hosting the model themselves on GPU-enabled hardware.
Written by urgent.news from Dev.to's report — not a translation of it. Machine-written — may contain errors; check the original before relying on it.
