Mycelis
Products Model hosting BYOK & Managed Keys Agents & Smart Routing Fine-Tuning (LoRA) Knowledge Bases & RAG MCP Tools API Gateway
Use Cases Compliance & Privacy Team rollout Cost Control Quick start with managed keys Your own assistants Fine-tuning with domain knowledge OpenClaw / OpenCode MCP Integration OpenAI compatibility
Resources Getting Started Documentation Blog Status
Models MCP Hub Pricing FAQ
Reference · 8 min min read

API Reference

Overview

The Mycelis API is fully compatible with the OpenAI Chat Completions API. Any SDK, tool, or integration that works with OpenAI works with Mycelis.

Official OpenAI endpoint and parameter docs: platform.openai.com/docs/api-reference

Authentication

Authorization: Bearer myk_...

Create API keys in Dashboard → API Keys.

Base URL

https://your-domain.com/api/proxy/v1

Use your Mycelis agent/deployment slug as model.

Chat Completions example

POST /api/proxy/v1/chat/completions
Authorization: Bearer myk_...
Content-Type: application/json

{
  "model": "my-agent",
  "messages": [
    { "role": "user", "content": "What is RAG?" }
  ]
}

Supported parameters

Parameter Type Description
model string Agent or deployment slug
messages array OpenAI-format message list
stream boolean Enable SSE streaming
temperature number Creativity (0–2)
max_tokens integer Max output length
top_p number Nucleus sampling
stop string/array Stop sequences

All additional parameters are forwarded to the underlying deployment/provider.

Error codes

Code Meaning
401 Missing or invalid API key
403 No workspace permission
404 Agent/deployment not found
400 Invalid request / missing provider key
503 Deployment not active