OpenCode gets stuck buffering and does not send API requests

[VineshF1]

Description

I'm facing an issue with OpenCode where prompts do not appear to trigger an API request to the server.

For example, when I type a simple prompt such as:

hello

or any other prompt, OpenCode keeps buffering indefinitely and I don't receive any response.

What I observed

• The prompt is accepted by OpenCode.
• The UI remains in a loading/buffering state.
• No corresponding API request appears to reach the server when I check the server/API logs.
• I tried switching between different models, but the behavior remains the same.
• Since the request does not seem to reach the server at all, this may be occurring before the model/API request is actually made.
• I'm not sure whether this is related to the device, account/session, configuration, or something else.

Troubleshooting attempted

• Tried a simple "hello" prompt.
• Tried multiple different prompts.
• Tried different models.
• Checked the API/server logs to verify whether requests were reaching the server.
• Confirmed that the issue persists across models.

Expected behavior

When I enter a prompt, OpenCode should send the corresponding API request and eventually display the model's response.

Actual behavior

The prompt remains stuck in a buffering/loading state indefinitely, and no API request appears in the server logs.

Additional information

I can provide relevant OpenCode logs, configuration details, or network/debugging information if needed.

The most notable observation is that the request appears to get stuck before an API request is sent to the server, rather than the server returning an error.

Plugins

No response

OpenCode version

2.0.18

Steps to reproduce

For example, when I type a simple prompt such as:

hello

or any other prompt, OpenCode keeps buffering indefinitely and I don't receive any response.

Screenshot and/or share link

No response

Operating System

Windows 10

Terminal

Windows Terminal

[github-actions]

This issue doesn't fully meet our contributing guidelines.

What needs to be fixed:

• Add concise reproduction steps and identify which client is affected (TUI, Desktop, or Web) and whether OpenCode is connected to a local or remote server.
• Replace “Latest” with the exact OpenCode version, and include the selected provider/model plus relevant sanitized logs or configuration so maintainers can determine where the request stops.

Please edit this issue to address the above within 2 hours, or it will be automatically closed.

---

This issue might be a duplicate of existing issues. Please check:

• Desktop client connected to remote server: AI model does not respond, stuck on "thinking" #17976: The Desktop client connected to a remote server stays on “thinking,” while the server logs show no incoming requests.
• Models Not Responding / Interface Freezes When Sending Any Prompt #41345: A Windows 10 report where the interface becomes unresponsive after sending prompts across models.

If you believe this was flagged incorrectly, please let a maintainer know.

[github-actions]

Thanks for updating the issue with reproduction details and system information!

[VineshF1]

I have solved this issue myself. opencode sucks, blocking other harness to use their free models