Give superpowers to your assistant

AIVAX facilitates the creation of intelligent agents connected to the world by providing functions and ready integration for most situations you need.

With AIVAX, you can add to your assistant:

RAG

Index thousands of documents intelligently to bring specialized knowledge to your assistant.

Connected to your API

Execute server-side tools, such as an MCP protocol, connected directly to your existing API.

Customization

Create custom assistants, instructed, parameterized, and fine-tuned according to your preference.

Reduce costs

Reduce inference costs by defining context limits, message generation, and message truncating.

Multimodal interpretation

Add interpretation of images, audios, and documents to models that do not have these characteristics.

Ready clients

Integrate your model into a ready-to-use chat client for web, Telegram, and WhatsApp.

Bring your own key

Do not limit yourself to our suppliers' models. Bring your own API-Key and use the models with your own pricing, limits, and all AIVAX features.

Complexity routing

Use complexity-based routing models, which select the best models for each use case, reducing costs and maintaining quality.

OpenAI Compatible

Bring your own OpenAI model and use your gateways created with an API fully compatible with the OpenAI Chat Completions standard.

Add pre-configured tools to your model

Our tools are highly configurable and happen on our end. Choose parameters between quality, context, and pricing for the tools.

Check out the ready-to-use tools for your assistant:

Web search

Provide internet access to your model so it can consult real-time information and specific data.

  • Choose between a summarized and complete search.
  • Choose how many search results to embed in the conversation.
Code evaluation

Code execution allows the assistant to perform precise calculations and solve problems that are better solved through algorithms or code.

URL context

Allow your assistant to access external links, download their content, and use it in the conversation.

Long-term memory

With this feature, the assistant takes important notes about the conversation and the user. With these notes, the assistant contextualizes itself with information across multiple conversations.

Image generation

Allow your assistant to generate images.

  • Choose how many images are generated per iteration.
  • Choose the image generation quality among low, medium, high, and very high.
  • Control whether the assistant can generate explicit content or not through image generation.
Chat client example
Launch in Minutes

Embed a sleek chat UI with a few lines of code. Manage context and quotas your way.

Simple, Transparent Pricing

No surprises. Pay only for what you actually use.

Pay-Per-Use
We pass provider token rates straight to you—no markups, no hidden fees.
Platform Fee
A small surcharge covers taxes and operational costs when you top up.

Connect Your AI to the World

Use JSON to call web search, run code, or fetch links. AIVAX handles the heavy lifting and returns exactly what you need.

// Request
> {
    "instructions": "Fetch today's SELIC and IPCA rates",
    "responseFormat": { "ipcaToday": 0, "selicToday": 0 }
}

// Response
< { "ipcaToday": 5.2, "selicToday": 14.0 }

Get Early Access

Become a beta user today and shape the future of AI gateways.

Join Beta Now