Detroit LLMDetroit LLM Sign in
Detroit LLM Platform

Start building
with Detroit LLM

Everything you need to integrate Detroit LLM into your applications. From first API call to production.

OpenAI compatible — use your existing SDK via /v1/chat/completions

Get started for free → View API Docs →
cURLPythonTypeScript
curl https://chat.khain.app/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-dt-..." \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello Detroit LLM"}]
  }'

Why Detroit LLM?

Everything you need — chat, API, key management and usage in one place.

Multiple models, one gateway

Switch models instantly in chat — DeepSeek for code, Qwen for general tasks, GLM for language — seamless.

OpenAI Compatible API

Endpoint /v1/chat/completions and /v1/models works with OpenAI SDK, Vercel AI SDK, LangChain instantly.

Chat + API Keys + Usage

Web chat, API key management, Requests/Tokens/Punchcard graphs and transparent Tier quotas.

Fast streaming

Go Gateway + SGLang for low latency streaming just like OpenAI.

MIKKUCN membership

Linked to YouTube Membership — auto-verified, or subscribe to Paid plan directly.

Stable LLM Gateway

Clear weekly/monthly Tier quotas — no need to top up multiple providers.

Multiple models — pick the right one for the job

Choose the model that excels at each task and switch seamlessly in one chat.

deepseek-v4-flashFast

DeepSeek V4 Flash — fast for everyday Q&A and high-volume tasks

Context1M context
Size304B
deepseek-v4-proFlagship

DeepSeek V4 Pro — most capable for reasoning, code and production

Context1M context
Size304B
qwen3.7-flashFast

Qwen 3.7 Flash — fast with thinking mode for stronger reasoning

Context128K context
Size—
qwen3.8-flashFast

Qwen 3.8 Flash — fast with thinking mode for stronger reasoning

Context128K context
Size—
z-image-turboImage

z-image-turbo — high-quality 1024×1024 image generation

Context1024×1024
Size—
glm-5.3Reasoning

GLM-5.3 — 1M context / max 128K output

Context1M context
Size128K max
glm-4.7-flashxReasoning

GLM-4.7-FlashX — fast and efficient

Context131K max
Size65K default

See all: GET https://chat.khain.app/v1/models

Get started in 3 steps

1

Sign in with Google

Auto-verifies YouTube Membership or choose a Paid plan

2

Get your API key

Create a key on the Keys page and use it with OpenAI SDK

3

Pick a model and chat

Switch models per task and monitor usage in real time

FAQ

Everything you need to know about Detroit LLM

What is Detroit LLM?

Detroit LLM is an AI Chat Platform and LLM Gateway that lets you chat with multiple AI models in one place with seamless model switching via OpenAI-compatible API.

Which models are supported?

Currently supports deepseek-v4-flash, deepseek-v4-pro, qwen3.7-flash, qwen3.8-flash, z-image-turbo, glm-5.3 and more. See the full list at /v1/models or the Models page.

Does it work with existing OpenAI code?

Yes. Just change base_url to https://chat.khain.app/v1 and use your Detroit LLM API key. Existing OpenAI SDK code works without changes.

How do I get started?

Click “Sign in with Google” — we verify your MIKKUCN YouTube Membership or Paid plan, then you get an API key and can start chatting or calling the API immediately.

How does pricing work?

Access is tier-based: YouTube members and Paid plans get different weekly/monthly token quotas. See details on the Upgrade plan after signing in.

Ready to try Detroit LLM?

Sign in with Google Join Discord