toki ca5eb4685c fix(edge): provider-pool thinking과 queue 정책을 정리한다 ... vLLM-MLX provider-pool에서 strict output이 thinking을 끄지 않도록 하고, queue_timeout_ms=0을 IOP queue timeout 없음으로 해석해야 한다. long reasoning/long-context 요청은 IOP queue timeout이 아니라 caller cancellation과 backend timeout 정책으로 제어한다.		2026-07-06 10:16:43 +09:00
..
bootstrap	feat: long context admission support - refactor edge server, add input estimator	2026-07-05 18:18:58 +09:00
configrefresh	feat: long context admission support - refactor edge server, add input estimator	2026-07-05 19:26:06 +09:00
controlplane	feat: edge runtime model queue and status dispatch updates	2026-07-05 20:01:27 +09:00
edgecmd	feat(edge): node resource model unification - update phase/SDD docs and edge service	2026-06-29 14:39:27 +09:00
edgevalidate	feat: routing status refresh and config validation updates	2026-06-30 09:47:34 +09:00
events
input	feat: long context admission support - refactor edge server, add input estimator	2026-07-05 18:18:58 +09:00
node	feat: openai-compatible tool call boundary hardening & knowledge tool optimization	2026-07-04 18:21:27 +09:00
openai	fix(edge): provider-pool thinking과 queue 정책을 정리한다	2026-07-06 10:16:43 +09:00
opsconsole	feat: reasoning_delta 이벤트 처리 및 출력 개선	2026-07-02 14:05:19 +09:00
service	fix(edge): provider-pool thinking과 queue 정책을 정리한다	2026-07-06 10:16:43 +09:00
transport	feat(runtime): workspace_root 제거와 tool 호환을 반영한다	2026-06-27 11:32:45 +09:00