Model Configuration Fixes

This cluster addresses various issues related to model configuration and caching in the OpenClaw framework.

15 PRs

fix models config memory Representative: #14744

#	Title	Author	Created	GitHub
3322	fix: merge provider config api into registry model	nulone	2026-01-28	View
9739	#9291 fix(models): preserve existing models in models.json when merging provi...	ximzzzzz	2026-02-05	View
12195	fix(agents): sync config fallback for lookupContextTokens cold-start race	mcaxtr	2026-02-09	View
13626	fix(model): propagate provider model properties in fallback resolution	mcaxtr	2026-02-10	View
14744	fix(context): key MODEL_CACHE by provider/modelId to prevent collision (#14708) rep	lailoo	2026-02-12	View
15581	Rename Qianfan label	ide-rea	2026-02-13	View
15632	fix: use provider-qualified key in MODEL_CACHE for context window lookup	linwebs	2026-02-13	View
16543	[AI-assisted] feat(usage): support cache-hit differentiated pricing	OwenJiong24	2026-02-14	View
17469	Improve unknown-model errors for provider/model misconfiguration	megahappyclaw	2026-02-15	View
17604	fix(context): use getAvailable() to prevent cross-provider model ID collision...	aldoeliacim	2026-02-16	View
19655	Fix/context window provider keying	pharasyte	2026-02-18	View
21998	fix(models): prioritize exact model-id match over fuzzy scoring (#21942)	lailoo	2026-02-20	View
22889	feat(talk): add provider-agnostic talk config contract	ngutman	2026-02-21	View
23136	fix: lookupContextTokens should handle provider/model refs	patchguardio	2026-02-22	View
23542	fix/hf inference	Josephrp	2026-02-22	View