3 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
michaelxerandmichaelxer	1fc7a5d35f	fix: fall back to /models probe when base URL returns 404 (#3205) ... _ping_endpoint() probes the bare base URL for non-Ollama endpoints. OpenAI-compatible servers like llama-swap return 404 on the /v1 prefix but 200 on /v1/models, causing endpoints to appear offline despite being fully functional. Add a /models fallback when the base URL returns a non-auth 4xx. Auth failures (401/403) are treated as definitive — probing /models would just repeat the same rejection. Fixes #3181 Co-authored-by: michaelxer <michaelxer@users.noreply.github.com>	2026-06-07 16:09:33 +02:00
michaelxerandmichaelxer	43c31ae008	fix: raise imaplib line limit for large mailboxes (#2895) ... Python's imaplib._MAXLINE defaults to 1 MB. Mailboxes with tens of thousands of messages exceed this on UID SEARCH ALL, crashing with 'got more than 1000000 bytes'. Set _MAXLINE to 50 MB after opening the connection so large mailboxes work without error. Fixes #2883 Co-authored-by: michaelxer <michaelxer@users.noreply.github.com>	2026-06-05 22:59:35 +02:00
michaelxerandmichaelxer	f07bf99236	fix: respect user round count in deep research (#2896) ... The STOP_PROMPT did not include the target round count, so the LLM could decide to stop after 2-3 rounds even when the user requested 8. Additionally, min_rounds was capped at 3 regardless of max_rounds. - Add max_rounds to STOP_PROMPT so the LLM knows the target - Change min_rounds from min(3, max_rounds) to max(2, max_rounds - 2) Fixes #2863 Co-authored-by: michaelxer <michaelxer@users.noreply.github.com>	2026-06-05 20:49:42 +02:00