API requests
Choose an endpoint and use cURL, Python or JavaScript.
OpenAI SDK and Codex base URL
https://zashmodel.com/v1Claude Code base URL
https://zashmodel.comClaude Code adds /v1/messages automatically. When sending a raw HTTP request, include /v1 exactly once in the full URL.
Use the protocol supported by the selected model. API compatibility applies to enabled endpoints and model capabilities; do not assume every provider feature is available.
| Method | Endpoint | Use case |
|---|---|---|
GET | /v1/models | List models available to your API key. |
POST | /v1/responses | Responses-compatible models and coding tools. |
POST | /v1/chat/completions | Chat Completions-compatible models and chat clients. |
POST | /v1/messages | Anthropic Messages-compatible models and Claude Code. |
Language
Endpoint
Install the OpenAI Python package, then run this file with the environment variables from Quick start.
python -m pip install openaiimport os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["ZASHMODEL_API_KEY"],
base_url=os.environ["ZASHMODEL_BASE_URL"],
)
response = client.responses.create(
model=os.environ["ZASHMODEL_MODEL"],
input="Hello!",
)
print(response.output_text)For images, video and other capabilities, first check the model detail page for the enabled endpoint, supported parameters and billing unit.