{{ message }}
Repository navigation
Add Responses API compatibility alongside chat/completions #232
Description
Problem
The proxy currently exposes /chat/completions, /v1/chat/completions, and related compatibility routes, but it does not expose OpenAI Responses API routes. This blocks models that are available only through Responses API, such as gpt-5.3-codex, even when the underlying GitHub Copilot account can use them.
Proposed change
- add
/responses - add
/v1/responses - forward requests to the upstream GitHub Copilot
/responsesendpoint - include verbose logging and token-count reporting similar to the existing chat-completions handler
Related PR
Implemented here: #230
Validation
- source build passes with
bun run build - locally verified
gpt-5.3-codexandgpt-5.4-minithrough/v1/responses - verified that
gpt-5.3-codexis not usable via/chat/completions, so Responses support is required
Activity
pmiscn commentedon May 7, 2026
0.7.0
--> POST /responses 404 0ms
<-- POST /responses
--> POST /responses 404 0ms
<-- POST /responses
--> POST /responses 404 0ms
<-- POST /responses
--> POST /responses 404 0ms
<-- POST /responses
--> POST /responses 404 0ms
johnib commentedon Jun 21, 2026
@toreleon the 404 is expected afaict - looks like 0.7.0 doesn't expose a
/responsesroute at all, so codex 404s before it even gets to model lookup. the codex models are responses-only so you can't reach them through chat/completions anyway.i maintain a fork that adds the responses passthrough (
/responses+/v1/responses), forwards straight to Copilot's native endpoint. been running it for codex daily, same one i linked in openai/codex#8361.you're welcome to give it a try :)