Mistral AI

3 models · 1 agent · 1 API · 1 app

Every change, in the Monday brief

Updates

All 9 changes

Mistral Large 4 and Microsoft Foundry

2 products from Mistral AI and Microsoft

Now available3d

Mistral Large 4 becomes available on Microsoft Foundry

Microsoft is the first cloud provider to offer Mistral Large 4, with private deployment options for Azure developers.

Mistral Large 4

Model at Mistral AI

New release3d

Mistral Large 4 releases open weights with three reasoning modes at $0.70 input

Mistral released the weights of the improved Mistral Large 4 under a modified MIT license. It now has Low, Medium and High reasoning modes, and pricing is $0.70 input, $0.07 cached input and $2.10 output per million tokens.

Before

Public preview with None reasoning mode

After

Open weights with Low, Medium and High (default) reasoning modes

GLM-5.3-Flash and Mistral AI Studio

2 products from Z.ai and Mistral AI

Now available1w

Z.ai GLM 5.3 Flash available on the Mistral API

Mistral added zai-glm-5-3-flash to its platform at $0.15 per million input tokens, $0.03 cached input and $0.50 output.

Models

Mistral Medium 3.5
Model
New release: Mistral Medium 3.5 released with open weights · Apr 28, 2026
Mistral Large 4
Model
Now available: Mistral Large 4 becomes available on Microsoft Foundry · 3d
Voxtral 3
Model
New release: Mistral releases the Voxtral 3 speech models in public preview · 1w

Agents

Vibe Code
Agent
Security: Mistral Vibe CLI 2.25.4 fixes six shell permission vulnerabilities · Sep 12, 2026

APIs

Mistral AI Studio
API
Now available: Z.ai GLM 5.3 Flash available on the Mistral API · 1w

Apps

Mistral Vibe
App
New release: Mistral launches Vibe, its unified agent · May 28, 2026