backend deprecated models check - 2026-10-23

Automated read-only check of the hardcoded backend model lists against https://ai-models.puter.work/. A model is listed only if none of its names (id + aliases) appear in the upstream ids, names or aliases.

chat/mistral (3)

• mistral-large-4
• mistral-large-2512
• zai-glm-5-3 Notes: upstream mistral has no "large" model at all. mistral-large-4 is a recent public-preview entry (release_date 2026-10-06) and may just not be in the upstream feed yet, so confirm before removing. zai-glm-5-3 is not a Mistral-served id, so it is likely stale or misplaced in the Mistral list.

tts/xai (1)

• xai-tts Notes: the upstream xai feed has no TTS entries, and this id is the service key still used by XAITTSProvider.ts. This is probably a false positive and not a real deprecation.

All other checked pairs have no deprecations: chat claude, openai, gemini, deepseek, moonshot, minimax, xai, zai and alibaba; image gemini, openai and xai; tts gemini and openai; video gemini. ai-video/providers/openai has no models file, so it was skipped.

Verify manually: zai. The endpoint lists only base GLM models. All 11 backend zai ids (including glm-5.3-flash and glm-5.3-flashx) did match upstream this run, so nothing is flagged, but the endpoint's coverage of -flash/-flashx/vision variants may change.

Generated by Claude Code