## [3.36.0]( (2026-10-21) ### Features * add Vertex AI support for the Responses API ([#4139]( ([3e63d33]( * **api:** add 30-minute cache-write usage counts ([#4111]( ([5589146]( * **api:** add compaction trigger IDs ([#4122]( ([0e5794d]( * **api:** add custom voice consent management ([#4135]( ([de6d70d]( * **types:** prepare native synchronous WebSocket option type ([#4118]( ([90c0bac]( ### Bug Fixes * **api:** accept canonical batch responses in strict mode ([#4106]( ([6eaa90e]( * clean up failed WebSocket setup and honor native options ([#4121]( ([b1a1a10]( * end sync WebSocket recovery when an attempt is interrupted ([#4129]( ([337769b]( * honor timed Translation receives during terminal cleanup ([#4132]( ([ca54146]( * preserve client headers in legacy Beta Realtime connections ([#4130]( ([8a2a2e0]( * preserve declared aliases during Pydantic v1 validation ([#4119]( ([8b1cac9]( * preserve Vertex API root for response lifecycle ([#4140]( ([c2b14f5]( * preserve WebSocket close across recovery and cancellation ([#4117]( ([4590ff1]( * **realtime:** honor native options for legacy beta sockets ([#4120]( ([8968560]( ### Chores * **api:** clarify cache-write token totals ([#4136]( ([55b8b8a]( * **api:** clarify project API key update requirements ([#4131]( ([0e6be9f]( * **api:** maintain voice consent compatibility ([#4137]( ([07abe06]( * **api:** use neutral moderation examples
openai-python v3.36.0 Adds Vertex AI Support
openai-python v3.36.0 adds Vertex AI support for the Responses API, plus cache-write usage counts, compaction trigger IDs, voice consent management, and multiple WebSocket fixes.