Part of 785+ curated AI tools on AISO
Mistral 3 logo

Mistral 3

Mistral's MoE flagship + edge model family — Apache 2.0, multimodal, reasoning

0
freemiumDR 86Apache 2.0 open weights — free to download and self-host. Available via Mistral API (pricing on mistral.ai/pricing). Accessible on Le Chat free and Pro plans.View full pricing →

Visit Mistral 3

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

About Mistral 3

Mistral's December 2025 model family: Mistral Large 3 (flagship sparse MoE, 41B active / 675B total parameters, Apache 2.0) plus the Ministral 3 series (dense 3B, 8B, 14B edge models). Mistral Large 3 is Mistral's first MoE since Mixtral, with multimodal capabilities and #2 ranking on LMArena for OSS non-reasoning models. Ministral 14B reasoning variant scores 85% on AIME '25. All released under Apache 2.0.

Key Features

Mistral Large 3: sparse MoE with 41B active / 675B total parameters
Ministral 3 series: dense 3B, 8B, 14B models optimized for edge inference
Ministral variants include base, instruct, and reasoning variants
Image understanding across 40+ languages (Ministral 14B)
Ministral 14B reasoning: 85% on AIME '25
Mistral Large 3 ranks #2 OSS non-reasoning on LMArena leaderboard
All models released under Apache 2.0 license
First Mistral MoE model since the seminal Mixtral series

Mistral 3 Pros & Cons

Pros

  • +Apache 2.0 license — truly free for commercial use, no restrictions
  • +MoE architecture in Mistral Large 3 delivers high quality at lower inference cost than dense 675B
  • +Ministral series covers the full spectrum: 3B for embedded devices up to 14B for edge servers
  • +85% AIME '25 for Ministral 14B reasoning — strong math reasoning at edge scale
  • +Multimodal image understanding built into Ministral models across 40+ languages
  • +#2 on LMArena OSS leaderboard validates real-world chat quality

⚠️ Cons

  • Mistral Large 3 at 675B total parameters requires significant multi-GPU infrastructure to self-host
  • Context window not disclosed in launch announcement — unclear for production planning
  • API pricing not detailed in announcement — check mistral.ai/pricing for current rates
  • Released December 2025, so less community tooling than Llama or older Mistral models

Who Is Mistral 3 Best For?

👤Developers needing a fully open (Apache 2.0) frontier-grade model
👤Edge deployments: Ministral 3B/8B for mobile, IoT, and on-device inference
👤Researchers who want a large MoE they can fine-tune without license restrictions
👤Teams building multilingual multimodal applications on constrained hardware

Tags

mistralllmapiopen-weightsmixture-of-expertsreasoningmultimodalself-hostededge modelapache 2.0
🏷️

Is this your tool?

Claim your listing to get a Featured badge, edit your description, and stand out from competitors. All plans include a permanent dofollow backlink to your site.

Claim Now →

📬 Get the best new AI tools delivered weekly

One concise email with fresh launches, trending picks, and featured standouts.

Alternatives to Mistral 3

View all Mistral 3 alternatives →