You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Documentation: Opper ChatOpenAI base_url example #6009
The supported-models reference shows OpenAI-compatible hosts (Qwen, ModelScope, Novita, PZERO) used through ChatOpenAI and base_url. Opper works the same way, but users have to find the base URL, the key variable and the model name format themselves.
Opper is an EU-hosted AI gateway with 700+ models from 50+ providers behind one OpenAI-compatible API. No new provider class is needed; ChatOpenAI with base_url is enough.
Disclosure: Opper is my startup. I have a docs-only PR ready that adds the section below.
Suggested Changes
Add an Opper subsection under "OpenAI-Compatible APIs", after PZERO:
Env: OPPER_API_KEY, get a key at https://platform.opper.ai. Use the /v3/compat base URL (not /v3/compat/chat/completions). Model names such as claude-sonnet-4-6 or gpt-5.4-mini cover every provider that serves the model. Models: https://opper.ai/models
Activity
docs: add Opper OpenAI-compatible provider example (browser-use#6009)