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
I’m seeing the same thing on macOS with OpenCode 1.4.3 and Ollama. The session reaches service=llm providerID=ollama modelID=gemma4:e4b ... stream and then stalls with no completion/error.
Direct calls to Ollama’s OpenAI-compatible endpoint work fine (/v1/chat/completions with gemma4:e4b, including tool calls), so this looks like an OpenCode streaming regression rather than an Ollama/model issue.
I also confirmed the same behavior in --pure mode, so it doesn’t seem plugin-related.
Activity
github-actions commentedon Apr 9, 2026
This issue might be a duplicate of existing issues. Please check:
Both are closed — it's possible this is a regression or a related but distinct case. Feel free to reference those issues for additional context.
RemeJuan commentedon Apr 10, 2026
I’m seeing the same thing on macOS with OpenCode 1.4.3 and Ollama. The session reaches service=llm providerID=ollama modelID=gemma4:e4b ... stream and then stalls with no completion/error.
Direct calls to Ollama’s OpenAI-compatible endpoint work fine (/v1/chat/completions with gemma4:e4b, including tool calls), so this looks like an OpenCode streaming regression rather than an Ollama/model issue.
I also confirmed the same behavior in --pure mode, so it doesn’t seem plugin-related.
github-actions commentedon Jul 3, 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.
dercodeKoenig commentedon Sep 14, 2026
still broken