Features
Vibe AI agent for long-horizon tasks with knowledge and tools
Vibe for Code: async code generation, CI/CD, legacy translation
Studio: build and deploy AI apps with agent orchestration
Forge: train and align custom models with RLHF
Compute: frontier-grade GPU clusters for training and inference
Search Toolkit: production search pipelines
Mistral OCR 4: state-of-the-art document intelligence
Voxtral TTS: text-to-speech synthesis
Physics AI for engineering simulation
Self-hosted, cloud, or multi-cloud deployment options
Enterprise-grade guardrails, evals, and observability
100+ enterprise connectors (beta)
Custom MCP connectors (beta)
Vibe voice mode for audio input and transcription
Task scheduling for automated workflows
Unified API for 400+ models across 70+ providers
Automatic failover routing when a provider goes down
Model Fusion: combine budget models to rival frontier performance
Subagent delegation: frontier models hand off tasks to cheaper workers
Speech/transcription APIs (May 2026)
Unified Image API with capability discovery across 30+ models (June 2026)
OpenAI SDK compatible out of the box
Server-side presets to survive model deprecation
Edge-deployed for minimal latency
Budget controls and spend limits
Credit-based pricing with no subscriptions
Model routing visualization for latency and cost
Private model support for enterprise deployments
Data policy-based routing and enforcement
MCP server for coding agents (June 2026)
