Integrate conversational AI, agents, and file intelligence into your own product.
Generate an API key from your developer dashboard, then send your first request:
curl https://api.gentlearrow.ai/v1/messages \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gentlearrow-pro",
"messages": [{"role": "user", "content": "Summarize this quarter'"'"'s report"}]
}'
Every response streams back as JSON with the assistant's reply, token usage, and the model used.
All requests require an API key passed as a bearer token in the Authorization header. Keys are scoped to your account and can be rotated or revoked from your dashboard at any time.
The Messages endpoint powers conversational requests. Pass a list of messages with alternating user and assistant roles, plus an optional system prompt to set behavior.
| Model | Best for | Context |
|---|---|---|
| gentlearrow-fast | Everyday chat, low latency | 32k tokens |
| gentlearrow-pro | Reasoning, coding, analysis | 128k tokens |
| gentlearrow-vision | Image understanding | 32k tokens |
Invoke a specialized agent by setting agent to research, coding, business, or creative in your request body. Agents carry a task through multiple steps and return a structured result.
Free accounts are limited to 60 requests per hour. Pro and Business plans raise this significantly \u2014 see your dashboard for your current limit and usage.
Errors return a JSON body with an error.type and error.message field, alongside the relevant HTTP status code (401 for auth issues, 429 for rate limits, 400 for malformed requests).