Mistral 3 Debuts: Open Multimodal Models from 3B to 675B

Mistral 3 is an Apache 2.0 open-source family: Ministral 3 (3B/8B/14B) and Mistral Large 3 — a 41B‑active/675B MoE. Multimodal, multilingual, optimized for edge and scalable deployment; available now across major plat...

Mistral 3 Debuts: Open Multimodal Models from 3B to 675B

TL;DR

  • Apache 2.0 open-source family: three dense Ministral 3 models (3B / 8B / 14B) plus a sparse Mistral Large 3 (41B active / 675B total); all models available now.
  • Mistral Large 3: sparse MoE trained from scratch on 3,000 NVIDIA H200; base and instruction checkpoints released, reasoning variant forthcoming; strong image understanding and multilingual conversation; LMArena #2 OSS non-reasoning, #6 OSS overall
  • Ministral 3: 3B/8B/14B sizes, each in base, instruct, and reasoning variants; image understanding and 40+ native languages; instruct models optimized for cost-to-performance and often produce fewer tokens; 14B reasoning reports 85% on AIME ’25.
  • Availability: Mistral AI Studio, Amazon Bedrock, Azure Foundry, Hugging Face, IBM WatsonX, OpenRouter, Fireworks, Unsloth AI, Together AI; coming soon on NVIDIA NIM and AWS SageMaker.

Mistral 3 arrives: an open, multimodal family from 3B to 675B parameters

Mistral AI has released Mistral 3, a new family of open-source models under Apache 2.0, spanning compact edge-friendly weights and a frontier-scale sparse model. The lineup includes three dense Ministral 3 models (3B, 8B, 14B) and a sparse mixture-of-experts, Mistral Large 3, with 41B active / 675B total parameters. All models are available today.

Mistral Large 3 — an open MoE trained at scale

Mistral Large 3 is a sparse MoE trained from scratch on 3,000 NVIDIA H200 GPUs. The release includes both base and instruction-fine-tuned checkpoints, with a reasoning variant to follow. After post-training, the model reaches parity with leading instruction-tuned open-weight models on general prompts, while also showing strong image understanding and multilingual conversation capabilities. On LMArena, Mistral Large 3 debuts at #2 in the OSS non-reasoning models category and #6 among OSS models overall: https://lmarena.ai/leaderboard/text

For deployment and experimentation, an optimized NVFP4 checkpoint is released via llm-compressor (https://github.com/vllm-project/llm-compressor), enabling efficient operation on Blackwell NVL72 systems and on a single 8×A100 or 8×H100 node using vLLM (https://github.com/vllm-project/vllm).

Industry co-optimization for training and inference

The release emphasizes co-design with industry partners. NVIDIA integration includes support for TensorRT-LLM (https://github.com/NVIDIA/TensorRT-LLM) and SGLang (https://github.com/sgl-project/sglang) for low-precision inference, plus Blackwell attention and MoE kernels and speculative decoding to improve long-context, high-throughput serving on GB200 NVL72 and similar platforms. Edge and local deployment paths are highlighted for DGX Spark (http://nvidia.com/en-us/products/workstations/dgx-spark/), RTX PCs and laptops (https://www.nvidia.com/en-us/ai-on-rtx/), and Jetson devices (https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/).

Ministral 3 — compact, multimodal, multilingual

The Ministral 3 series targets edge and local inference scenarios with 3B, 8B, and 14B sizes. Each size ships in base, instruct, and reasoning variants, and all variants include image understanding and support for 40+ native languages. The Ministral instruct models emphasize cost-to-performance, often producing fewer generated tokens while matching or exceeding comparable models. For tasks prioritizing accuracy, the Ministral 14B reasoning variant reports 85% on AIME ’25.

Availability, customization, and documentation

Mistral 3 is available now via multiple platforms, including:

Model documentation and technical resources:

For organizations pursuing tailored deployments, Mistral AI offers custom model training services: https://mistral.ai/solutions/custom-model-training

Community and next steps

Model checkpoints, instruction-tuned variants, and deployment tooling are provided to support research, engineering, and enterprise integration. Further resources and community channels are listed on the documentation pages and platform listings linked above.

Original source: https://mistral.ai/news/mistral-3

Continue the conversation on Slack

Did this article spark your interest? Join our community of experts and enthusiasts to dive deeper, ask questions, and share your ideas.

Join our community