# Mistral publishes Ministral 3 14B Instruct FP8 quantized weights

> Mistral AI published the FP8 quantized variant of the Ministral-3-14B-Instruct model on Hugging Face under Apache 2.0. Uses the vllm library.

| | |
|---|---|
| **Tool** | Mistral AI |
| **Version** | — |
| **Kind** | release |
| **Published** | 2025-10-31 |
| **Observed** | 2026-08-11 |
| **Significance** | 1/5 |
| **Breaking** | no |
| **Categories** | model-support |

> **Note:** this classification is below our confidence threshold and is pending human review.

## What changed


- Model published at `mistralai/Ministral-3-14B-Instruct-2512`
- Library: vllm
- Base model: mistralai/Ministral-3-14B-Base-2512 (quantized)
- Weight format: FP8
- License: Apache 2.0
- Tags: arxiv:2601.08584


## Sources

- [hf_model](https://huggingface.co/mistralai/Ministral-3-14B-Instruct-2512) — retrieved 2026-08-11


## Community

_No curated reactions recorded. Facts and community takes are kept in separate layers
and never blended._

---
Canonical: https://changelogs.info/mistral/mistral-publishes-ministral-3-14b-instruct-fp8-quantized-weights
Entity: https://changelogs.info/mistral
Event ID: `evt_2025-10-31_mistral_mistralai-ministral-3-14b-instruct-2512`
Licence: event synthesis © changelogs.info, CC BY 4.0. Linked sources belong to their vendors.