Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Mistral AI - Mixture of experts - verified

Run Mistral Large 3 675B Instruct 2512 locally

Mistral Large 3 675B Instruct 2512 is a 675B mixture-of-experts (41B active) model from Mistral AI with a 262K-token context window, released under Apache 2.0 (commercial use is allowed); the smallest listed download is 407.0 GB, the Q4_K_M GGUF build from unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF.

Official weights: mistralai/Mistral-Large-3-675B-Instruct-2512 - modalities: text, image

Total params
675B
Active params
41B
Context
262K
Smallest download
407.0 GB

Fits in

Smallest download at most 85% of memory

None of the tiers here. The smallest listed download is 407.0 GB, more than 85% of 256 GB.

Downloads

The official weights and every quantization listed for this model, with file sizes from Hugging Face.

Mistral Large 3 675B Instruct 2512 downloads and file sizes
DownloadFormatSizeRepo
Official weightsFP8681.5 GBmistralai/Mistral-Large-3-675B-Instruct-2512
Q4_K_MSmallest listedGGUF407.0 GBunsloth/Mistral-Large-3-675B-Instruct-2512-GGUF
Q8_0GGUF715.7 GBunsloth/Mistral-Large-3-675B-Instruct-2512-GGUF

GB is decimal: 1 GB = 10^9 bytes. Add memory for the context and the operating system on top of the file size.

Run it

Copied verbatim from the page linked under each command. Server commands from model cards often assume a multi-GPU node; check the flags before you run them.

vLLM

Terminal
vllm serve mistralai/Mistral-Large-3-675B-Instruct-2512 \
  --max-model-len 262144 --tensor-parallel-size 8 \
  --tokenizer_mode mistral --config_format mistral --load_format mistral \
  --enable-auto-tool-choice --tool-call-parser mistral

Source: huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512

Not sure which runtime to use? Compare local LLM runtimes by platform, API and features.

License

License
Apache 2.0apache-2.0
Commercial use
Commercial use: YesCommercial use is allowed.
Gated download
No. The files download directly.

Worth knowing

The HF API reports no safetensors parameter total for this repo, so the 675B total and 41B active figures come from the model card, and the Ollama entry is cloud only.

Released (per the announcement).

API pricing and specsHosted prices per 1M tokens for Mistral Large 3 675B Instruct 2512, if you would rather not run it yourself.

FAQ

How big is the Mistral Large 3 675B Instruct 2512 download?

The official FP8 weights in mistralai/Mistral-Large-3-675B-Instruct-2512 are 681.5 GB. The smallest listed download is 407.0 GB, the Q4_K_M GGUF build from unsloth/Mistral-Large-3-675B-Instruct-2512-GGUF. Plan for more memory than the file size: the context (KV cache) and the operating system need room too.

Can I use Mistral Large 3 675B Instruct 2512 commercially?

Mistral Large 3 675B Instruct 2512 is released under Apache 2.0, and commercial use is allowed. The license text is at https://www.apache.org/licenses/LICENSE-2.0.

How much memory do I need for Mistral Large 3 675B Instruct 2512?

Its smallest listed download is 407.0 GB, more than 85% of 256 GB, so it needs a multi-GPU server or a machine with more memory than the largest tier here.

What is the context length of Mistral Large 3 675B Instruct 2512?

262,144 tokens. A longer context needs more memory for the KV cache, so you can run with a shorter one when memory is tight.

Sources

Last verified from the pages below.

Newsletter

New open weights, every week

One email per week. New model releases, open-weight drops and what they mean for what you build.

Keep exploring

Best local AI models

Our curated picks by hardware tier, with benchmarks and the reasoning.

Model API pricing and specs

First-party prices per 1M tokens when you would rather call an API.

AI benchmark leaderboards

Sourced scores, for when you need to compare quality.

Setup guides

Point your coding tools at the model you picked.