TUI crashes immediately with {"data": {},"errors": [],"success": false} after upgrade to v1.14.42 (Distrobox/container environment)

[knas4ever]

Description

Description:
After auto-upgrading from a previous version to v1.14.42, the TUI fails to start and immediately exits with:
json{
"data": {},
"errors": [],
"success": false
}
Only one log line is produced before the crash:
INFO 2026-05-09T18:28:24 +148ms service=default version=1.14.42 args=["--print-logs"] process_role=main run_id=... opencode
Environment:

OS: Bazzite (Fedora-based, immutable) with Distrobox container (Ubuntu)
Terminal: Ptyxis (GNOME), TERM=xterm-256color, COLORTERM=truecolor
opencode runs inside a Distrobox container (ai-box) accessed via podman exec
Node.js: v18.19.1
LM Studio local provider via opencode-lmstudio@latest plugin

What works:

opencode run "message" works perfectly — server starts, model responds, tools work
opencode web works — web interface accessible on port 4096
opencode models lmstudio works
Previous version of opencode worked fine in the same environment today before the auto-upgrade at 18:35 local time

What doesn't work:

opencode (TUI mode) — crashes immediately
The crash happens regardless of --port, TERM=xterm, plugins disabled (--pure), or minimal config

Root cause hypothesis:
The TUI client makes a GET /config request to the server on startup and receives HTTP 400. This causes an immediate crash. The same config endpoint works fine in opencode web. This may be a regression in how the TUI initializes or handles the config response in v1.14.42.
Relevant log excerpt from opencode run (which works) showing the 400:
INFO service=default http.method=GET http.url=/config http.status=400 logSpan.http.span.2=2ms Sent HTTP response
Config:
json{
"$schema": "https://opencode.ai/config.json",
"plugin": [
"opencode-lmstudio@latest",
"superpowers@git+https://github.com/obra/superpowers.git"
],
"autoshare": false,
"model": "lmstudio/qwen3.6-27b",
"provider": {
"lmstudio": {
"npm": "@ai-sdk/openai-compatible",
"name": "LM Studio (local)",
"options": {
"baseURL": "http://127.0.0.1:1234/v1"
},
"models": {
"qwen3.6-27b": {
"limit": {
"context": 200000,
"output": 4096
}
}
}
}
}
}
Workaround: opencode web or opencode run "message" until fixed.

Plugins

No response

OpenCode version

No response

Steps to reproduce

No response

Screenshot and/or share link

No response

Operating System

No response

Terminal

No response

[github-actions]

This issue doesn't fully meet our contributing guidelines.

What needs to be fixed:

• OpenCode version field is empty — please fill in the exact version (e.g. v1.14.42)
• Steps to reproduce field is empty — please provide a minimal set of steps to reproduce the crash
• Operating System field is empty — please fill in your OS (you mentioned Bazzite/Fedora-based with Distrobox/Ubuntu, add that here)
• Terminal field is empty — please fill in your terminal (you mentioned Ptyxis)

The description body contains much of this information, but the structured template fields are required so maintainers can triage efficiently.

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

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

[kitlangton]

Hey, can you try opencode upgrade to pull v1.14.43? It shipped about an hour after v1.14.42 with #26550, which fixed a /config shape regression that looks like exactly what you're hitting.

If it still crashes on v1.14.43, paste the GET /config 400 response body and the server logs from the crashing TUI start so we can see which validation is failing.

[github-actions]

This issue has been automatically closed because it was not updated to meet our contributing guidelines within the 2-hour window.

Feel free to open a new issue that follows our issue templates.