6 Commits

Author	SHA1	Message	Date
Bartok9	5bb78ea7ed	fix(model): preserve LM Studio '@' quant suffixes in model name resolution ... stripModelProfileSuffix() in providers.ts naively truncated model names at the first '@', discarding quant variants like @iq3_xxs, @iq4_xs, @q4_k_xl that LM Studio uses to distinguish quantization levels. This caused two user-facing bugs (fixes #71474): 1. /model lmstudio/qwen3.6-27b@iq3_xxs → 'model not allowed: lmstudio/qwen3.6-27b' 2. API requests sent truncated model name → LM Studio picked a random quant Changes: - Replace the naive indexOf('@') strip in providers.ts with splitTrailingAuthProfile() which already handles quant suffixes - Extend the quant-suffix regex (q\d+...) to also match importance- quantization tags (iq3_xxs, iq4_xs, ...) via i?q\d+ pattern - Add tests for @iq* quant suffixes and auth-profile-after-iq combos	2026-04-26 01:41:08 +01:00
foxtrot026	02c092e558	fix(model-ref): recompute suffix after @YYYYMMDD + add @8bit test	2026-04-06 13:18:59 +01:00
foxtrot026	5208a85afe	fix(model-ref): treat LM Studio/Ollama @q*/@4bit suffixes as model-id	2026-04-06 13:18:59 +01:00
Alix-007	2c579b6ac1	fix(models): preserve @YYYYMMDD version suffixes (#48896) thanks @Alix-007 ... Co-authored-by: Alix-007 <267018309+Alix-007@users.noreply.github.com> Co-authored-by: frankekn <frank.ekn@gmail.com>	2026-03-18 13:20:06 +08:00
Peter Steinberger	e4e5d9c98c	fix(model): land #30932 auth-profile @ parsing for /model (@haosenwang1018) ... Landed from contributor PR #30932 by @haosenwang1018. Co-authored-by: haosenwang1018 <haosenwang1018@users.noreply.github.com>	2026-03-02 03:05:03 +00:00
Peter Steinberger	4b259ab81b	fix(models): normalize trailing @profile parsing across resolver paths ... Co-authored-by: Vincent Koc <vincentkoc@ieee.org> Co-authored-by: Marcus Castro <mcaxtr@gmail.com> Co-authored-by: Brandon Wise <brandonawise@gmail.com>	2026-02-26 14:34:15 +01:00