mARTin-B78 ccccdf820a feat: graph-decoded streaming, prebuffer_ms and cancel-on-disconnect (v6.13) ... - Fixed-window CUDA-graph codec decode for streaming (TTS_GRAPH_DECODE=0 disables) - Per-request chunk_size (2-24) on /v1/audio/speech - prebuffer_ms (0-10000) to hold back audio before sending - Stop generation on client disconnect - Recommend --max-seq-len 2048 for the streaming container Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>		2026-10-04 00:11:12 +02:00
..
openai_server.patch	feat: graph-decoded streaming, prebuffer_ms and cancel-on-disconnect (v6.13)	2026-10-04 00:11:12 +02:00