Mistral AI
Open and portable generative AI for devs and businesses
Mistral AI is a Paris-based AI company building frontier open and commercial language models. They offer comprehensive AI solutions including inference APIs, developer tools (Studio), AI agents (Vibe), model training (Forge), and compute infrastructure, combining open-source leadership with commercial-grade enterprise solutions.
At a glance · prices collected Nov 21, 2026
What's good about Mistral AI
- Strong open-source model ecosystem with Apache 2.0 licensing
- Comprehensive AI platform beyond just inference APIs
- European company with EU data residency options
- Excellent code generation with dedicated Codestral model and Vibe coding agent
- Full-stack AI solutions from training to deployment
- OpenAI-compatible API for easy migration
- Advanced developer tooling with Studio and multi-platform agent access
Mistral AI LLM API prices
Pay-per-token pricing, per 1M tokens.
Prices last updated November 21, 2026
| Model | Input/1M | Output/1M |
|---|---|---|
| $0.100 | $0.100 | |
| $0.150 | $0.500 | |
| $0.150 | $0.600 | |
| $0.150 | $0.150 | |
| $0.200 | $0.200 | |
| $0.300 | $0.900 | |
| $0.300 | $1.20 | |
| $0.700 | $2.10 | |
| $1.40 | $4.40 |
Mistral AI pricing by model
Input, output, and batch rates, plus alternatives, for one model at a time.
Billing & terms
- Vibe Subscription Plans
- Free, Pro, Team, and Enterprise tiers with varying usage limits and features
- API Pay-per-token
- Per million token pricing with separate input, cached input, and output rates; OCR billed per page and audio per minute or character
- Batch and Priority tiers
- Standard, Batch, and Priority inference tiers with separate rates, plus regional inference options
- Voice Agents
- Voice Agents API billed per minute of conversation
- Free tier
- Rate-limited free access for experimentation
- Enterprise deployments
- Custom models, private deployments, and dedicated infrastructure
- Regions
- EU (France) primary hosting with global availability and in-region inference options. On-premises and VPC deployment available. Azure, AWS Bedrock, and Google Vertex AI deployment options for data residency requirements
- Support
- Documentation, Discord community, Vibe chat interface, email support, and enterprise support plans
Watch out for
- Newer platform with less third-party ecosystem compared to OpenAI
- Complex product suite may require learning curve for new users
- Enterprise features still maturing compared to hyperscale providers
- Full platform capabilities require higher-tier subscriptions
Features
- Mistral Model Family
- Access to Mistral Large 4, Mistral Small 4, Ministral 3 (14B, 8B, 3B), and specialized models like Codestral, OCR 4.1, and Voxtral TTS 3
- Third-Party Open-Weight Models
- Hosted third-party models served through the Mistral API, including Z.ai GLM 5.3 and DeepSeek V4.1 Flash
- Vibe AI Agent
- Unified agent for long-horizon productivity and coding tasks, available on web, mobile, CLI, and IDE
- Studio Developer Platform
- Complete AI development environment with API playground, agent orchestration, and deployment tools
- Forge Model Training
- Train, align, and evaluate custom AI models with enterprise domain adaptation
- Mistral Compute
- Frontier-scale infrastructure for training and inference with dedicated GPU clusters
- Open-Weight Models
- Open-weight models including Mistral Large 4, Mistral Small 4 (Apache 2.0), Shieldstral safety models, and Timestral time series forecasting models for self-hosting
- Function Calling
- Native tool use and function calling across all commercial models
- JSON Mode
- Structured output with guaranteed valid JSON responses
- Multimodal Capabilities
- Vision support, OCR, speech transcription including realtime (Voxtral Transcribe), speech generation (Voxtral TTS), and document intelligence
- Search Toolkit
- Production search pipelines with 100+ enterprise connectors and MCP support
Getting started
Step 1
Create an account
Sign up at console.mistral.ai to access Vibe and Studio
Step 2
Choose your starting point
Use Vibe for AI agent tasks or Studio for API development
Step 3
Generate API key
Create an API key from Studio for programmatic access
Step 4
Install tools
Install Vibe CLI, SDK (mistralai), or use web interfaces
Step 5
Start building
Make your first API call or run tasks in Vibe
Tools that work with Mistral AI
Coding agents, gateways and other software that can use Mistral AI, listed by name. All tools
AnythingLLMAI chat apps
ClineAI coding agents
Cloudflare AI GatewayLLM gateways and routers
HeliconeLLM gateways and routers
JanAI chat apps
Kilo CodeAI coding agents
LangfuseLLM observability and evals
LangSmithLLM observability and evals
LibreChatAI chat appsVia an OpenAI-compatible endpoint
LiteLLMLLM gateways and routers
MLflowExperiment tracking
Open WebUIAI chat appsVia an OpenAI-compatible endpoint
PortkeyLLM gateways and routers
Where this page's facts come from
- LLM API prices
- docs.mistral.ai · last collected Nov 21, 2026
- Profile
- Reviewed Mar 14, 2026
- Method
- How we collect and check prices
Frequently Asked Questions
Which models does Mistral AI serve?
Check the pricing table above for the models Mistral AI currently serves and their per-token rates.
How do I get started with Mistral AI?
Create an account, Choose your starting point, Generate API key, Install tools, Start building
What are Mistral AI's main advantages?
Mistral AI's main advantages include: Strong open-source model ecosystem with Apache 2.0 licensing, Comprehensive AI platform beyond just inference APIs, European company with EU data residency options, Excellent code generation with dedicated Codestral model and Vibe coding agent, Full-stack AI solutions from training to deployment, OpenAI-compatible API for easy migration, Advanced developer tooling with Studio and multi-platform agent access.
What are Mistral AI's limitations?
Mistral AI's main limitations include: Newer platform with less third-party ecosystem compared to OpenAI, Complex product suite may require learning curve for new users, Enterprise features still maturing compared to hyperscale providers, Full platform capabilities require higher-tier subscriptions.
Compare Providers
Find the best prices for the same GPUs and models from other providers · All Mistral AI alternatives, ranked →
All Mistral AI Comparisons
The weekly price report
One email every Monday with the week's notable GPU and LLM API price moves, plus occasional product news. No spam.