Commit Graph
68 Commits
Author SHA1 Message Date
Kan Wu 0f6761b54f [sgl-router] Add SGLang-compatible DeepSeek V4.1 Flash rendering (#40532) 2026-09-21 18:35:09 +08:00
0abb251a20 [sgl-router] Match DeepSeek V4 rendering to SGLang (#40530)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-21 17:56:41 +08:00
Kan WuandClaude Fable 5.1 2016f5e7a1 [sgl-router] Add Kimi-K3 rendering with SGLang parity (#40390)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-21 17:38:49 +08:00
630b1ef322 [sgl-router] Fix reorg admission proxy test build after BucketResolver::new (#40537)
Co-authored-by: Kan Wu <wukanustc@gmail.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-21 16:42:29 +08:00
Kan WuandClaude Fable 5.1 a9871012ac [sgl-router] refactor - generalized admission policy definitions (#40271)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-21 16:26:30 +08:00
Kan WuandClaude Fable 5.1 fcb080bd40 [sgl-router] refactor - cache-aware policy (#40366)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-21 13:53:57 +08:00
Kan WuandClaude Fable 5.1 4027740569 [sgl-router] refactor - session-aware policy (#40379)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-20 17:22:36 -07:00
Kan WuandClaude Fable 5.1 aedda8377e [sgl-router] refactor - layout BucketResolver, Bucket, EngineGroup and implement PowerOfTwo (#40241)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-20 16:57:07 -07:00
Kan WuandClaude Fable 5.1 4a9dc5c4af [sgl-router] refactor - move policy-required states under src/state (#40272)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-20 16:49:23 -07:00
Kan Wu c610c40399 [sgl-router] refactor - config and organize CLI options (#39867) 2026-09-20 19:40:04 +08:00
Kan WuandCursor 671630abf1 [sgl-router] refactor - main startup logic (#39861)
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-20 16:55:34 +08:00
Kan Wu 99d53fe0c2 [sgl-router] refactor - chat_completions() into modules (#39848) 2026-09-20 13:51:25 +08:00
aed3fb1cdd [Router] Log every request at one site; derive its outcome from the final status (3/3) (#39465)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-19 03:11:00 +08:00
4e0b56c811 [Router] Treat an upstream 503/429 as backpressure, not a breaker fault (2/3) (#39464)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-19 00:46:29 +08:00
6de4666e43 [Router] Derive error status from a failure class; preserve the worker's status (1/3) (#39463)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-18 16:32:45 +08:00
3bf243d6a9 [Router] Shard the cache-aware KV tree by chain root (#39167)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-18 15:00:17 +08:00
72d9419bef [Router] Sample k random candidates for the min-load fallback (--min-load-choices) (#39170)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-18 03:09:09 +08:00
e4cbb28ea1 [router] Improve SGLang chat render parity (#39133)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-17 23:53:40 +08:00
Kan WuandClaude Fable 5.1 3ce7e2a29f [sgl-router] Render chat prompts with dynamo-render (#38983)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-17 15:57:20 +08:00
b02e16a895 [Router] Fleet-wide sampling contract 3/3: splice injection without re-serializing (#39002)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 09:59:48 -07:00
3e03879f68 [Router] Drain readiness before SIGTERM shutdown so k8s deregisters the pod first (#39016)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 09:59:26 -07:00
Kan Wu 279339f113 [sgl-router] Share model-file discovery for chat formatters (#39485) 2026-09-16 09:36:11 -07:00
Kan WuandClaude Fable 5.1 ad94978adf [sgl-router] Rename chat encoder to chat formatter (#39459)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-16 08:46:11 -07:00
Kan WuandClaude Fable 5.1 e41026f434 [sgl-router] Forward input_ids only for string content; count tokenize errors only when forwardable (#39458)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-16 22:49:41 +08:00
8baeded6f3 [Router] Pin to the prefix owner when the whole fleet is queueing (--saturation-queue-floor) (#39169)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 16:16:24 +08:00
54f0d72adf [Router] Fleet-wide sampling contract 2/3: enforce and inject per request (#39001)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 00:29:33 -07:00
7835f1de9a [Router] Add the shutdown-drain configuration surface (#39015)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 00:26:50 -07:00
e2d56bbbfc [Router] Bound the e2e worker memory budget so prefill graph capture stops OOMing (#39713)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 00:13:09 -07:00
d634320e48 [Router] Count open HTTP exchanges until their response body finishes (#39014)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 00:04:38 -07:00
afde31a2f5 [router] Speak cleartext h2c on both edges: serve it inbound, forward it outbound (#39006)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 22:30:31 -07:00
c1f5b4736a [Router] Add --worker-queue-limit: stop sending cache-affinity traffic to a queueing worker (#39168)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 13:24:48 +08:00
5f17e3a75f [Router] Fleet-wide sampling contract 1/3: the config surface (#39000)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 21:39:06 -07:00
bbef93e22a [Router] Add a KV storage-tier coverage row to the Grafana dashboard (4/4) (#39111)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 17:09:11 -07:00
63845a1bb2 [router] Resolve a wire protocol per worker at registration (#39004)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 16:54:37 -07:00
fb91baedab [Router] Real-GPU e2e coverage for storage-tier-aware cache routing (3/4) (#39110)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 15:48:51 -07:00
3c48c1e967 [Router] Expose the KV storage-tier stream and tree occupancy on /metrics (2/4) (#39109)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 13:30:05 -07:00
0e9b6bf8d8 [Router] Extract worker selection into policies::selection (no behavior change) (#39322)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-15 15:49:31 +08:00
ishandhanani 3f871a246c feat(agent sessions): attribute stored KV cache blocks to sessions (#37482)
Signed-off-by: Ishan Dhanani <ishandhanani@gmail.com>
2026-09-14 21:34:49 -07:00
45b511b2ef [Router] Honor KV-event storage tiers in the cache-aware tree (1/4) (#39108)
Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-14 21:18:15 -07:00
Kan WuandClaude Fable 5.1 e89d8facab [sgl-router] Prepare dynamo-render dependencies (#39457)
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
2026-09-14 19:08:16 -07:00
Kan Wu f539c1fc65 [sgl-router] Stream outcome observability for 2xx SSE streams (#38737) 2026-09-13 19:46:44 -07:00
ishandhanani ddc1df1203 fix(router): simplify bucket context limit check (#39259)
Signed-off-by: Ishan Dhanani <ishandhanani@gmail.com>
2026-09-13 12:27:59 -07:00
Vincent Gao 8e7deb329e [Router] Preserve global cache affinity with bucket routing (#38814)
Signed-off-by: Vincent Gao <vincentbo@linux.alibaba.com>
2026-09-11 17:41:27 +08:00
Kan WuandClaude Fable 5 6f481ad0e3 [router] Raise chat body cap to 32 MB for multimodal payloads (#38735)
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
2026-09-09 22:14:49 -07:00
5bebe7a033 [Router] Add bucket-aware policy domains and native cache indexing (#38108)
Signed-off-by: Vincent Gao <vincentbo@linux.alibaba.com>
Co-authored-by: inkcherry <mingzhi.liu@amd.com>
Co-authored-by: yangbodong22011 <13137470+yangbodong22011@users.noreply.github.com>
2026-09-06 19:47:51 +08:00
ecd97de1fc [Router] Add load-aware prefill admission and bounded policy proposals (#37843)
Signed-off-by: Vincent Gao <vincentbo@linux.alibaba.com>
Co-authored-by: Kangyan Zhou <zky314343421@gmail.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
2026-09-05 11:53:26 +08:00
Vincent Gaoandinkcherry 54cadad151 [Router] Add composable scoring and eligibility policies (#37731)
Co-authored-by: inkcherry <mingzhi.liu@amd.com>
2026-09-04 00:01:41 +08:00
Alex NailsandAlison Shao 28262c20df [CI][RFC] Replace black-jupyter with ruff-format (#37210)
Co-authored-by: Alison Shao <a.shao@wustl.edu>
2026-09-02 19:46:08 -07:00
360d10d6bc [Feature] Add process-local in-memory KV indexer and Router integration (#33370)
Co-authored-by: Wu, Yutong <yutong.wu@amd.com>
Co-authored-by: TianDi101 <ditian12@amd.com>
Co-authored-by: Zhangheng <hzh0425@apache.org>
2026-08-20 10:45:35 +08:00
Cheng Wan c70c7d72a8 config: the readback and the resolving view say what they are (#35027) 2026-08-17 16:18:19 -07:00