[CaT-Carlo19]

Description

Document R-Mem using GoModel for an OpenCode Go LLM while keeping embeddings in LM Studio. The recipe avoids a duplicated /v1 path, separates the gateway key from the upstream Go key, and warns that R-Mem's global OPENAI_API_KEY also affects its embedding client.

Checked the example against R-Mem's current request construction and GoModel's provider-qualified model selection. The TOML parses and the docs diff has no whitespace errors. This is source-checked documentation, not an end-to-end live GoModel deployment claim. The compatibility example deliberately uses a Chat Completions model rather than assuming Muse Spark uses that protocol.

The recipe preserves Go's real coding-session and workload requirements and explicitly states that local storage/embeddings do not make LLM requests local. No runtime behavior changes.

[coderabbitai]

Important

Draft PR not reviewed

Draft PRs are not automatically reviewed by default.

• Trigger a manual review

To automatically review draft PRs, update your CodeRabbit configuration:

reviews:
  auto_review:
    drafts: true

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[SantiagoDePolonia]

@CaT-Carlo19 What is this AI spam about?

[SantiagoDePolonia]

The project you try to advertise wasn't updated for 6 months. Closing this PR and reporting as a spam