7 commits

Author	SHA1	Message	Date
tin-berri	4307b34aca	fix: omit thinking.type=disabled for always-on thinking Claude models (#37510)	2026-08-21 10:27:26 -07:00
mateo-berri	b39a339b7d	fix(vertex_ai): apply regional endpoint uplift to cost tracking	2026-08-19 15:21:06 -07:00
mateo-berri	be594f5984	feat(guardrails): count bedrock guardrail cost against spend and budgets ... Price ApplyGuardrail usage units recorded by PR #37225 with a new bedrock/guardrails entry in the model cost map (regional override via bedrock/{region}/guardrails), add the per-request guardrail_cost to the standard logging payload's response_cost and CostBreakdown, surface it in the x-litellm-response-cost header, and bill blocked requests through the failure hook so key and team budgets see what AWS bills	2026-08-18 14:16:07 -07:00
mateo	6517c1dc06	fix(cost): support cache_creation_input_token_cost in tiered pricing and make tier selection all-or-nothing ... Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>	2026-08-13 02:18:10 +00:00
mateo-berri	00e8691064	ci: enforce format assertions so calendar-impossible deprecation dates fail validation	2026-07-28 16:11:22 -07:00
mateo-berri	3d01c39d00	ci: tighten deprecation_date pattern to reject impossible months and days	2026-07-28 15:24:24 -07:00
mateo-berri	4556dfa930	ci: publish a generated JSON schema for model_prices_and_context_window.json	2026-07-27 12:11:17 -07:00