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
OpenCode Desktop 1.18.2 sends messages but models never reply after update #37255
Since updating to OpenCode Desktop 1.18.2 on Windows, I can no longer receive any response from the model.
I can type and send a message, but OpenCode remains stuck and no answer is displayed. The API key is correctly configured and was working before the update. I also checked that the configured API key is valid.
There is no visible error in the interface. The request appears to be sent, but the API or model never returns a response.
This issue started immediately after updating OpenCode Desktop.
Plugins
No
OpenCode version
1.18.2 desktop
Steps to reproduce
Steps to reproduce
Open OpenCode Desktop 1.18.2 on Windows.
Configure a valid API key in the provider settings.
Open a new conversation.
Select a model.
Enter any message, for example: “Hello”.
Send the message.
The message is submitted, but no response is ever received from the model.
Expected behavior
The selected model should return a response normally.
Actual behavior
The message is sent, but the conversation stays without any model response. There is no useful error message shown in the application.
If your issue is the same as one of these, please add a comment there with any additional details (e.g. your debug zip) rather than opening a new report. If it is distinct, feel free to keep this open.
Error from provider (Console): Model deepseek-v4-flash-free is not supported on the full model list. Use GET /inference/openai/v1/models to list available models.
Reacted by Tobiloba Oyelekan, Mykhajlo Kindratyshyn and romain-l
Opencode sometimes suddenly stops working—specifically, failing to automatically proceed to the next round of interaction after executing a shell command that should have finished quickly. In such cases, one must manually press the stop button and enter "continue" to get the agent moving again. This is quite frustrating.
I use 1.18.2 desktop (deployed on remote ubuntu servers).
Activity
github-actions commentedon Jul 16, 2026
This issue might be a duplicate of existing issues. Please check:
If your issue is the same as one of these, please add a comment there with any additional details (e.g. your debug zip) rather than opening a new report. If it is distinct, feel free to keep this open.
amcique commentedon Jul 16, 2026
yeah they fked up pretty bad:
Error from provider (Console): Model deepseek-v4-flash-free is not supported on the full model list. Use GET /inference/openai/v1/models to list available models.
JohnSpirit commentedon Jul 17, 2026
Opencode sometimes suddenly stops working—specifically, failing to automatically proceed to the next round of interaction after executing a shell command that should have finished quickly. In such cases, one must manually press the stop button and enter "continue" to get the agent moving again. This is quite frustrating.
I use 1.18.2 desktop (deployed on remote ubuntu servers).
github-actions commentedon Sep 25, 2026
To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.