Repository Issue Activity (beta)

#273 Requests ending in a role:"system" message get rejected with invalid_message_role

#272 role:system messages inside `messages` are mapped to assistant -> upstream 400 invalid_message_role

#270 Support Auto Max as default model for internal Auto rollout

#266 Is this a stale repo?

#186 claude still stuck the openning screen can't use

#264 Claude code's auto mode doesn't work for Opus 4.6

#262 Remove the HTTP request towards `https://aur.archlinux.org`

#46 [Claude Code] Frequently stuck on tool use feature.

#217 HTTP 400: context_management: Extra inputs are not permitted (Claude Opus 4.6)

#231 Add Responses API compatibility alongside chat/completions

#232 Add Responses API compatibility alongside chat/completions

#160 The account that is already logged in cannot be logged out

#242 为什么模型列表没有Claude系列？我的copilot订阅是有的

#255 How to use multiple accounts?

#121 ALL ANTHROPIC MODEL CANT USE

#218 如果用docker启动,会更便捷

#177 message: 'model gpt-5.1-codex-max is not accessible via the /chat/completions endpoint', code: 'unsupported_api_for_model' } }

#240 Fork with Opus-4.7 / GPT -5.5/ GPT-5.4 / GPT-5.3-codex / GPT-5.4-mini / gemini-3.1-pro / gemini-3-flash support, feedback welcome

#251 What are the counting rules for premium interactions?

#222 'max_tokens' is not supported with this model

#237 createChatCompletions crashes with TypeError when using GPT-5.4 (should return 400, returns 500 instead)

#233 THIS REPOSITORY IS NO LONGER MAINTAINED — PLEASE USE THE ACTIVE FORK INSTEAD

#117 gpt-5-codex model endpoint support + Codex CLI documentation

#228 Why my github premium request analytics stop counting my usage?

#221 Timed out Error