OpenAI-compatible
An OpenAI-compatible API for the code you already have.
- POST /v1/chat/completions
- Streaming and tools
- 4 models
Setup
Two lines to switch
Change base_url or baseURL and paste your sk- key. Sign up needs only an email; API access unlocks after your first purchase.
from openai import OpenAI
client = OpenAI(
base_url="https://api.claudech.com/v1",
api_key="sk-...",
)
response = client.chat.completions.create(
model="claude-opus-5-5",
messages=[{"role": "user", "content": "Hello"}],
)Compatibility
What works
Same request shapes as the OpenAI Chat Completions API for the features below.
Chat Completions
POST /v1/chat/completions with the usual messages array, model field and optional parameters.
Streaming
Set stream: true and read server-sent events from the same endpoint.
Tool calling
Send tools and tool_choice in the OpenAI tools format; responses include tool_calls when the model invokes a function.
Image input
User messages can include image content parts alongside text on every model.
Model list
GET /v1/models returns the public model ids you can pass to Chat Completions.
1M-token context
Every model accepts up to one million tokens of context in a single request.
Embeddings, image generation, audio, files and the OpenAI Responses API are not offered on Claudech.
Models
Model rates
Prices are per million tokens, the same on every plan.
| Model id | Max output | Input / 1M | Cached input / 1M | Output / 1M |
|---|---|---|---|---|
claude-opus-5-5Fast and capable | 66K tokens | $0.50 | $0.010 | $1.90 |
claude-fable-5-1Advanced reasoning | 131K tokens | $0.50 | $0.010 | $1.90 |
claude-opus-5Deep reasoning | 131K tokens | $0.50 | $0.010 | $1.90 |
claude-sonnet-5Fast responses | 33K tokens | $0.50 | $0.010 | $1.90 |
See SDK examples, the migration guide, tool integrations, or Cursor with Claudech.
FAQ
Frequently asked questions
Is Claudech affiliated with OpenAI?
No. Claudech is an independent service and is not affiliated with, endorsed by or sponsored by OpenAI. It implements the public Chat Completions request format so existing OpenAI SDK code keeps working when you change the base URL and key.
Which base URL should the OpenAI SDK use?
https://api.claudech.com/v1, with no trailing slash. That is the OpenAI-format root; paths such as /chat/completions are appended by the SDK or your HTTP client.
How do I authenticate?
Send your sk- key as Authorization: Bearer sk-... or as x-api-key: sk-.... Both work on every endpoint. API access unlocks after your first token purchase.
How am I billed?
Every request is billed per token from your prepaid balance at the model rates in the table below. The Usage page lists each request with its model, tokens and cost. Requests that fail before the model produces output are not billed. Purchased tokens never expire.
What is not supported in the OpenAI format?
Embeddings, image generation, audio, files and the OpenAI Responses API are not offered. Chat Completions (including streaming, tools and image input), the model list and account endpoints are.
Does streaming and tool calling work?
Yes. Set stream: true for server-sent events on Chat Completions, and send tools and tool_choice in the same shape the OpenAI API expects. Every model supports tool calling and a 1M-token context window.
Point your OpenAI client at Claudech
Create an account, buy any token package to unlock API keys, and set the base URL to https://api.claudech.com/v1.