canonical: https://jentic.com/apis/api.respan.ai/respan-ai

# Respan AI API

Jentic publishes the only available OpenAPI specification for the Respan AI API, keeping it validated and agent-ready. The Respan AI API runs chat completions against large language models through a single OpenAI-compatible endpoint. It accepts a conversation of messages along with the model and generation parameters and returns the model's reply, so an agent can call an LLM without wiring a provider SDK.

## For AI agents

Run chat completions against large language models through an OpenAI-compatible endpoint. Send a conversation of messages with the model and generation parameters and receive the model's reply.

## Scope

Does not handle model fine-tuning, embeddings, or file storage. Use for running chat completions against large language models only.

## Capabilities

- Run chat completions against large language models
- Send a conversation of messages and receive the model's reply
- Select the model and generation parameters for each request

## Use cases

### LLM-Backed Agent

An AI agent uses Respan AI as its language-model backend, sending a conversation of messages and reading back the completion to drive its next step. Through Jentic the agent discovers the chat completions operation by intent and calls it directly, so it can reason and generate text without a provider SDK wired into the app.

Example prompt: Send a conversation and return the model's completion for the agent's next step

### Content Generation

Generate drafts, summaries, or replies by sending a prompt and reading the model's response. This lets an agent produce copy on demand while the model and parameters stay under the caller's control, keeping output consistent across requests.

Example prompt: Summarize a block of text into three bullet points using a chosen model

### Structured Extraction

Turn unstructured text into structured fields by prompting the model to return a defined shape. The agent sends the text and instructions and parses the completion, extracting values without a dedicated parser for each format.

Example prompt: Prompt the model to extract a name and date from a sentence and return them as JSON

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/chat/completions` | Run a chat completion |

## Key resources

- **Chat completions** — Run chat completions against large language models

## Why Jentic

- **Setup:** Wiring the Respan AI API by hand means handling its bearer-token auth and the OpenAI-compatible request shape yourself. Through Jentic you install once, import Respan AI from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** Respan AI exposes one chat completions operation, so scoping is about whether the agent may call it, not which record it touches. You allow that operation explicitly, and every call it makes is logged.
- **Credential handling:** Your Respan AI token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'run a chat completion', and Jentic returns the Respan AI chat completions operation with its input schema so the agent calls the model without browsing the reference docs.

## Related APIs

- **OpenRouter** — Multi-model LLM gateway compared with Respan AI's completions endpoint
- **OpenAI** — OpenAI's own models, whose chat completions shape Respan AI mirrors
- **Anthropic** — Claude models available alongside Respan AI's completions

## FAQ

### What authentication does the Respan AI API use?

The Respan AI API authenticates with a bearer token in the Authorization header, per its OpenAPI spec. Through Jentic the token is stored encrypted by your own Jentic One instance and injected at call time, so it never enters the agent's prompt or logs.

### Can I run a chat completion with the Respan AI API?

Yes. Call POST `/chat/completions` with a list of messages and the model you want. The endpoint is OpenAI-compatible, so a conversation and model parameters return the model's generated reply.

### What are the rate limits for the Respan AI API?

The OpenAPI spec does not specify rate limits. Check the Respan AI documentation at https://api.respan.ai before sending high-volume completion traffic.

### Why is there no official OpenAPI spec for the Respan AI API?

Respan AI does not publish its own OpenAPI description. Jentic maintains this specification from the documented chat completions endpoint and keeps it validated so agents can call the API reliably.

### Is there a Respan AI MCP server?

You don't need an MCP server to give your agent the Respan AI API. Jentic connects it directly from the API Directory: import it, store your token once, and your agent calls the chat completions operation.

### Can I limit what my agent is allowed to do with the Respan AI API?

Yes. Respan AI exposes a single chat completions operation, so a rule controls whether the agent may call it at all, and every call it makes is logged.

### How do I run a chat completion with the Respan AI API through Jentic?

Search Jentic for 'run a chat completion', and it returns the POST `/chat/completions` operation with its input schema so your agent calls the model directly. To run it on your own infrastructure, install Jentic One from its GitHub repo.