docs: add Eden AI (EU, OpenAI-compatible) provider example#2052
docs: add Eden AI (EU, OpenAI-compatible) provider example#2052MVS-source wants to merge 1 commit into
Conversation
Eden AI is an EU-based, GDPR-compliant OpenAI-compatible gateway. Adds a worked config.toml example under the openai provider type, verified live against https://api.edenai.run/v3 (openai/gpt-4o-mini and mistral models).
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 9af042e9db
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| api_key = "sk-xxxxx" | ||
| ``` | ||
|
|
||
| ### Example: Eden AI (EU gateway, OpenAI-compatible) |
There was a problem hiding this comment.
This adds a new Eden AI subsection only to docs/en/configuration/providers.md, while the mirrored docs/zh/configuration/providers.md still jumps from openai directly to openai_responses. That leaves the two locale pages with different section structure and means Chinese readers will not see this new provider example, so please add the corresponding Chinese section in the same change.
Useful? React with 👍 / 👎.
1 participant
What
Adds a worked configuration example for Eden AI under the
openaiprovider type inconfiguration/providers.md.Eden AI is a French, EU-based, GDPR-compliant gateway that exposes 700+ models from many vendors (OpenAI, Anthropic, Mistral, Google, …) behind a single OpenAI-compatible key. Since it speaks the OpenAI Chat Completions protocol, it plugs into Kimi Code with
type = "openai"and a custombase_url— no code change needed. This just documents the exactconfig.tomlso users don't have to reverse-engineer the endpoint and theprovider/modelnaming.Verification
The example was tested live against
https://api.edenai.run/v3/chat/completions:openai/gpt-4o-mini→ HTTP 200mistral/mistral-large-latest(EU model) → HTTP 200Notes
/provider(code), before writing any code.