Repository navigation
Closed
Labels
bugSomething is broken or behaving incorrectly.Something is broken or behaving incorrectly.questionFurther information is requestedFurther information is requested
Description
Before submitting
- I searched existing issues and did not find a duplicate.
- I included enough detail to reproduce or investigate the problem.
Area
apps/desktop
Steps to reproduce
start app wthe npx t3 serve
then add it to the t3 app local
get the url and code put it and connect
send a message to my throw t3 to the remote and the model that i have is my self hosted ollama qwen3.6
here are the configs on the remote
cat ~/.config/opencode/opencode.json
{
"$schema": "https://opencode.ai/config.json",
"provider": {
"ollama": {
"npm": "@ai-sdk/openai-compatible",
"name": "CHPC Ollama",
"options": {
"baseURL": "*********************"
},
"models": {
"qwen3.6:27b": {
"name": "Qwen 3.6 27B",
"reasoning": true,
"tool_call": true,
"variants": {
"thinking": {
"reasoningEffort": "high"
},
"low": {
"reasoningEffort": "low"
}
}
}
}
},
"local": {
"npm": "@ai-sdk/openai-compatible",
"name": "Local H200 Ollama",
"options": {
"baseURL": "****************"
},
"models": {
"gemma4:31b": {
"name": "Gemma 4 31B",
"reasoning": true,
"tool_call": true
},
"qwen3.6:27b": {
"name": "Qwen 3.6 27B Local",
"reasoning": true,
"tool_call": true
}
}
}
},
"model": "ollama/qwen3.6:27b",
"agent": {
"primary": {
"description": "Primary coding agent",
"model": "local/qwen3.6:27b",
"temperature": 0.2
},
"builder": {
"description": "Fast builder",
"model": "ollama/qwen3.6:27b",
"temperature": 0.1
},
"planner": {
"description": "Planning agent",
"model": "local/gemma4:31b",
"temperature": 0.3
}
}
}
as you can see i have no chatgpt setup here
{
threadId: '2a57968e-a00e-4074-8dde-1a2a962d6863',
cwd: '/',
cause: 'TextGenerationError: Text generation failed in generateThreadTitle: Codex CLI command failed: 2026-05-25T03:22:41.757975Z ERROR codex_login::auth::manager: Failed to refresh token: 401 Unauthorized: {\n' +
' "error": {\n' +
' "message": "Your refresh token has already been used to generate a new access token. Please try signing in again.",\n' +
' "type": "invalid_request_error",\n' +
' "param": null,\n' +
' "code": "refresh_token_reused"\n' +
' }\n' +
'}\n' +
'2026-05-25T03:22:41.758072Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:41.797871Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:41.880306Z ERROR codex_models_manager::manager: failed to refresh available models: unexpected status 401 Unauthorized: Provided authentication token is expired. Please try signing in again., url: https://chatgpt.c
**********n' +
'2026-05-25T03:22:41.897677Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'OpenAI Codex v0.121.0 (research preview)\n' +
'--------\n' +
'workdir:' +
'model: gpt-5.4-mini\n' +
'provider: openai\n' +
'approval: never\n' +
'sandbox: read-only\n' +
'reasoning effort: low\n' +
'reasoning summaries: none\n' +
'session id: 019e5d27-fb58-72c3-a3ff-9c80bfb25503\n' +
'--------\n' +
'user\n' +
'You write concise thread titles for coding conversations.\n' +
'Return a JSON object with key: title.\n' +
'Rules:\n' +
"- Title should summarize the user's request, not restate it verbatim.\n" +
'- Keep it short and specific (3-8 words).\n' +
'- Avoid quotes, filler, prefixes, and trailing punctuation.\n' +
'- If images are attached, use them as primary context for visual/UI issues.\n' +
'\n' +
'User message:\n' +
'hello there\n' +
'warning: Codex could not find bubblewrap on PATH. Install bubblewrap with your OS package manager. See the sandbox prerequisites: https://developers.openai.com/codex/concepts/sandboxing#prerequisites. Codex will use the vendored bu
bblewrap in the meantime.\n' +
'2026-05-25T03:22:41.934874Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:41.971831Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:41.991896Z ERROR rmcp::transport::worker: worker quit with fatal: Transport channel closed, when UnexpectedContentType(Some("text/plain; body: {\\n \\"error\\": {\\n \\"message\\": \\"Provided authentication to
ken is expired. Please try signing in again.\\",\\n \\"type\\": null,\\n \\"code\\": \\"token_expired\\",\\n \\"param\\": null\\n },\\n \\"status\\": 401\\n}"))\n' +
'2026-05-25T03:22:42.006850Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:42.230619Z ERROR codex_api::endpoint::responses_websocket: failed to connect to websocket: HTTP error: 401 Unauthorized, url: wss://chatgpt.com/backend-api/codex/responses\n' +
'2026-05-25T03:22:42.265361Z ERROR codex_login::auth::manager: Failed to refresh token: Your access token could not be refreshed because your refresh token was already used. Please log out and sign in again.\n' +
'2026-05-25T03:22:42.434419Z ERROR codex_api::endpoint::responses_websocket: failed to connect to websocket: HTTP error: 401 Unauthorized, url: wss://chatgpt.com/backend-api/codex/responses\n' +
i have a codex running there with an expired token but it is defaulting to using chatgpt while i have chosen a different model in the app
Expected behavior
choose the correct model on the remote also a better way to add the remotes since adding them was hard
Actual behavior
prompt failing and not responsive after 16 minutes no feedback and thread is stuck at working
can't stop the thread so i have to delete it
Impact
Blocks work completely
Version or commit
No response
Environment
No response
Logs or stack traces
Screenshots, recordings, or supporting files
No response
Workaround
No response
Metadata
Metadata
Assignees
Labels
bugSomething is broken or behaving incorrectly.Something is broken or behaving incorrectly.questionFurther information is requestedFurther information is requested
Activity
bielarusajed commentedon May 26, 2026
Check "Text generation model" in settings — I think it is defaulted to GPT-5.4-mini via Codex, that's why it is trying to use it.
juliusmarminge commentedon Jul 20, 2026
The quoted call is
generateThreadTitle, which uses the separate Text generation model setting rather than the OpenCode conversation model. Please confirm whether a conversation turn itself invokes Codex after setting the text-generation model explicitly; if so, attach logs showing that separate path.\n\nsaleemalharir1 commentedon Jul 20, 2026
thanks looks like an update to the latest version solved the issue