# Mistral Large

Mistral has announced expansions to its AI infrastructure and product offerings, including plans for up to 1 GW of European AI compute capacity by 2030, new regional inference endpoints with priority tiers, and the release of Mistral Small 4 as a unified hybrid model. The company also introduced Vibe, a unified AI agent for work and coding, and an open source Search Toolkit for building production

Status: provisional
Last verified: 2026-08-29T13:40:05.000Z

## Evidence

- (single_source) Mistral Priority Tier is now in public preview.
  - supports https://mistral.ai/news/regional-inference-open-models-new-compute
    > Our new Mistral Priority Tier, now in public preview

- (single_source) Mistral Small 4 is a hybrid model that supports both text and image inputs.
  - supports https://mistral.ai/fr/news/mistral-small-4
    > Mistral Small 4 is a hybrid model optimized for general chat, coding, agentic tasks, and complex reasoning. Its architecture supports both text and image inputs, making it versatile for a wide range of applications.

- (single_source) Mistral plans to build up to 1 GW of European AI compute capacity by 2030.
  - supports https://mistral.ai/news/regional-inference-open-models-new-compute
    > With plans to build up to 1 GW of capacity by 2030, Mistral aims to provide a scalable, sovereign AI infrastructure that retains value and control for users.

- (single_source) Search Toolkit is released by Mistral in public preview as an open source composable framework for building production search pipelines.
  - supports https://mistral.ai/news/search-toolkit
    > Today, we're releasing Search Toolkit in public preview. Search Toolkit is a composable framework for building production search pipelines for AI applications... Search Toolkit is open source and runs wherever your infrastructure does.

- (single_source) Mistral is introducing regional endpoints and priority tiers for in-region inference.
  - supports https://mistral.ai/news/regional-inference-open-models-new-compute
    > The company is expanding open model access, introducing regional endpoints and priority tiers, and forming a coalition to secure long-term European AI compute capacity.

- (single_source) Vibe is Mistral's unified AI agent for both work and coding, available in free, Pro, Team, and Enterprise plans.
  - supports https://mistral.ai/news/vibe-agent
    > Vibe is Mistral's unified AI agent for both work and coding... Available in free, Pro, Team, and Enterprise plans

- (single_source) Mistral Small 4 is the first Mistral model to unify capabilities of Magistral for reasoning, Pixtral for multimodal, and Devstral for agentic coding into a single model.
  - supports https://mistral.ai/fr/news/mistral-small-4
    > Mistral Small 4 is the first Mistral model to unify the capabilities of our flagship models, Magistral for reasoning, Pixtral for multimodal, and Devstral for agentic coding, into a single, versatile model.

- (single_source) Vibe runs on flagship Mistral models optimized for reasoning, agentic tasks, tool calls, and coding.
  - supports https://mistral.ai/news/vibe-agent
    > Vibe runs on flagship Mistral models optimized for reasoning, agentic tasks, tool calls, and coding.

## Timeline

- 2026-08-29T13:31:20.000Z: Mistral plans up to 1 GW of European AI compute capacity by 2030 Mistral introducing regional endpoints and priority tiers for in-region inference Mistral Priority Tier now in public preview Mistral Small 4 unifies Magistral, Pixtral, and Devstral capabilities into a single model Mistral Small 4 is a hybrid model supporting text and image inputs Vibe is Mistral's unified AI agent for work and coding across free, Pro, Team, and Enterprise plans Vibe runs on flagship Mistral models optimized for reasoning, agentic tasks, tool calls, and coding Mistral released Search Toolkit as open source composable framework in public preview