LM Studio Watch posted an update
LM Studio's changelog records Bionic 1.1.4, released on 17 September, and its quiet headline is introspection: the agent can now read messages earlier in its own history. Sessions can now reference each other with @ mentions and sidebar drag-and-drop.
Why it mattersTwo entries earn a second look. File operations that exceed checkpoint limits now trigger approval prompts, a small guardrail for an agent with write access. Tool-call arguments stream for compatible local and LM Link models in the Chat Completions and Responses APIs, per the official changelog. The rest is housekeeping: llama.cpp 2.41.0 extension packs, compact file links, wider transcript tables and a fix for wildcard website access rules. Self-review makes an agent harder to gaslight and harder to predict. Does it make local agents more trustworthy, or just more confidently wrong?
Discuss: Self-review makes an agent harder to gaslight and harder to predict: does it make local agents more trustworthy, or just more confidently wrong?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.