MI

Mistral Large 4 1T A52B NVFP4

MistralGGUF by ggml-orgUpdated 2026-10-29
Download on Hugging Face
Parameters1050B
Context1024K
Layers3
Architecturedflash
Smallest quant4.2 GB
Quick answer

Mistral Large 4 1T A52B NVFP4 needs about 4.8 GB of memory at BF16 (a 4.2 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 4.2 GB (BF16) to 619.4 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 0.2 GB.

Mistral Large 4 1T A52B NVFP4 – 1 trillion‑parameter multimodal model

Mistral Large 4 1T A52B NVFP4 is a 1 trillion‑parameter multimodal model with a 1M‑token context window. It supports text and image input, reasoning, tools, and native function calling. The model is quantized with NVFP4 for reduced memory usage.

Good at

  • Long‑context reasoning
  • Multimodal text‑image tasks
  • Tool‑enabled interactions

Limits

  • Requires Blackwell GPU for optimal performance
  • No accuracy regressions noted but no other quantizations available
  • License requires commercial agreement for MSA

Best for

  • Enterprise LLM services
  • Research on multimodal reasoning
  • Large‑scale text generation

Who it’s for. Developers building high‑capacity multimodal applications who have access to Blackwell GPUs and can comply with the Mistral Ecosystem Open‑Weight License. The model is suitable for commercial deployments that can obtain a separate commercial license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
BF164.2 GB 4.8 GB
Fits Get file →
—619.4 GB (13 parts) 620.0 GB
Too big Get file →

For 8 GB VRAM, use the BF16 quantization file (4.21 GB, ~4.8 GB at 8 k context). If more memory is available, the 619.41 GB unknown quant can be used, but it requires ~620 GB at 8 k context and is split into 13 shards.

What GPU do you need for Mistral Large 4 1T A52B NVFP4?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 8 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 16 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 24 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 32 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).
  • 48 GB VRAM: the best quant that fits is BF16 (4.2 GB file, about 4.8 GB in use).

Which Mistral Large 4 1T A52B NVFP4 quant should you choose?

BF16 (4.2 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedflash
Parameters1050B
Layers3
Trained context1,048,576 tokens
KV cache (fp16)6 KB per token

Source

Quantized byggml-org
FamilyMistral
Licenseother
Downloads (HF)1,179
Last updated2026-10-29

Frequently asked questions about Mistral Large 4 1T A52B NVFP4

How much VRAM does Mistral Large 4 1T A52B NVFP4 need?

About 4.8 GB at BF16 with an 8K context window. That is the 4.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Mistral Large 4 1T A52B NVFP4 on an 8 GB GPU?

Yes, with BF16: the 4.2 GB file needs about 4.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Mistral Large 4 1T A52B NVFP4 GGUF?

It depends on the quant: BF16 is 4.2 GB, file is 619.4 GB.

What context length does Mistral Large 4 1T A52B NVFP4 support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.2 GB at 32K.

Where can I download Mistral Large 4 1T A52B NVFP4 GGUF?

The files are on Hugging Face in the ggml-org/Mistral-Large-4-1T-A52B-NVFP4-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Mistral Large 4 1T A52B NVFP4 use?

Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Mistral Large 4 1T A52B NVFP4 on 8 GB VRAM?

Yes, the BF16 quantization file is 4.21 GB and requires about 4.8 GB at 8 k context, fitting within 8 GB VRAM.

Does this model support vision?

Yes, it accepts image input and produces text output.

What license applies to this model?

It is released under the Mistral Ecosystem Open‑Weight License, a modified MIT license; commercial use requires a separate license.

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More Mistral models

All Mistral →

Other large models

All large models →