Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.
Why it stands out
What to watch
Release timeline
release
Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.
View sourceNearby alternatives
Mistral Small Creative is an experimental small model designed for creative writing, narrative generation, roleplay and character-driven dia
Devstral 2 is a state-of-the-art open-source model by Mistral AI specializing in agentic coding. It is a 123B-parameter dense transformer mo
The largest model in the Ministral 3 family, Ministral 3 14B offers frontier capabilities and performance comparable to its larger Mistral S
A balanced model in the Ministral 3 family, Ministral 3 8B is a powerful, efficient tiny language model with vision capabilities.
Mistral launched Mistral: Mistral Large 3 2512
Compare
vs Google: Gemini 3.1 Flash Lite Preview
Side-by-side pricing, context, and capabilities
vs Qwen: Qwen3.5-35B-A3B
Side-by-side pricing, context, and capabilities
vs Qwen: Qwen3.5-27B
Side-by-side pricing, context, and capabilities
vs Qwen: Qwen3.5-Flash
Side-by-side pricing, context, and capabilities
vs Qwen: Qwen3.5 Plus 2026-02-15
Side-by-side pricing, context, and capabilities
vs Qwen: Qwen3 Coder Next
Side-by-side pricing, context, and capabilities