dipoleml
Docs menu

Getting started

raven App

API

DCode

Account

Resources

Docs

Introduction

Everything you need to build on the raven model family: the API, DCode, the raven app, and your account.

Dipole ML, Inc. builds the raven model family: frontier models engineered and served in Canada, with the API, tools, and infrastructure around them. This documentation covers the four surfaces developers work with:

  • The API: an OpenAI-compatible chat completions API at api.dipoleml.com, serving the raven model family today and whatever comes next. Streaming, vision, a reasoning channel, and pay-as-you-go billing with prepaid credits.
  • DCode: the agentic coding environment. It reads your codebase, plans the change, edits files, runs commands, and reports back.
  • The raven app: chat on the web at chat.dipoleml.com, with desktop and mobile apps.
  • Your account: one sign-in for keys, plans, usage, and billing.

#Two models

Every surface runs the same two models. Raven Flash is the fast one: production code, responsive agents, and high-volume pipelines. Raven Max is the deep one: long-horizon reasoning, careful analysis, and decisions you have to defend. Both read text and images, call tools, and stream their reasoning.

curl
curl https://api.dipoleml.com/v1/chat/completions \
  -H "Authorization: Bearer $RAVEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Raven Flash",
    "messages": [{"role": "user", "content": "Write a haiku about the northern lights"}]
  }'

#Where to go next