Commit Graph
100 Commits
Author SHA1 Message Date
Ke Bao 693adabff7 Fix Mamba2Metadata dropping has_mamba_track_mask (#26877) 2026-06-01 22:03:59 +08:00
Ke Bao cdd06011a1 Make unified tree SWA hicache tests faithful to write-through backup (#26870) 2026-06-01 16:08:57 +08:00
Ke Bao 972fbf7711 Skip flaky mamba extra_buffer disagg test (#26838) 2026-05-31 15:59:03 +08:00
Ke Bao e7b12fe6fa Fix stale forward_metadata leak in DP attn unpadded idle batch (#26313) 2026-05-25 16:04:00 -07:00
Ke Bao b13d3d18c6 Refactor HiCache stack dispatch into strategies (#26295) 2026-05-26 00:06:17 +08:00
Ke Bao e4378ff37f [UnifiedRadixCache] Fix HiCache load back start node (#25088) 2026-05-14 13:18:12 +08:00
Ke Bao 6be1a45d99 Fix swa component host hit (#25085) 2026-05-12 21:03:39 +08:00
Ke Bao d7f4761a48 [PD] Refactor hybrid state transfer (#24932) 2026-05-12 13:16:54 +08:00
Ke Bao d9cb38012e Update MiMo V2.5 cookbook image to nightly (#24983) 2026-05-11 22:04:31 +08:00
Ke Bao 62edbc37c4 [Doc] Add rerun-test slash command usage (#24979) 2026-05-11 21:30:01 +08:00
Ke Bao 9c03171f6c Fix gb envs in deployment guide (#24977) 2026-05-11 21:06:02 +08:00
Ke Bao 36ba60ad49 Register SWA unit tests under unit/mem_cache (#24974) 2026-05-11 20:21:05 +08:00
Ke Bao 59faf986b2 [PD] Unify dsv4 dispatch with swa (#24888) 2026-05-10 22:01:13 +08:00
Ke Baoandhzh0425 eb5f0fbeef Support swa HiCache for unified radix cache (#23391)
Co-authored-by: hzh0425 <hzh0425@apache.org>
2026-05-06 22:19:25 +08:00
Ke Bao 6dd7aebb36 Minor scheduler fixes (#24359) 2026-05-05 02:01:23 +08:00
Ke Bao aea527afdc Fix swa chunk req deferred (#24318) 2026-05-04 14:52:15 +08:00
Ke Bao 50fc2c9e23 Fix hybrid swa chunked prefill oom (#23174) 2026-04-21 10:46:45 +08:00
Ke Bao 3c0a6c6987 Add page_size and SWA coverage to unified radix cache bench test (#22815) 2026-04-14 23:58:05 +08:00
Ke Bao 9f9e0231bb Refactor unified radix cache UT into parameterized test suite (#22812) 2026-04-14 22:34:33 +08:00
Ke Bao bc1bfbf607 Fix swa input length limitation (#22597) 2026-04-12 16:03:35 +08:00
Ke Bao e77bfba24d Fix NCCL AllGather hanging issue for Qwen3 Next MTP (#22458) 2026-04-10 11:40:54 +08:00
Ke Bao 021bd77c91 Add skills for debugging hanging issues (#22463) 2026-04-10 01:37:44 +08:00
Ke Bao be42fbbbd7 Support HTTP2 server (#21700) 2026-04-08 00:42:52 +08:00
Ke Bao fae90abf6e Move ring test to nightly (#22267) 2026-04-07 21:56:39 +08:00
Ke Bao b95324d19c Update test coverage report (#22190) 2026-04-06 20:59:38 +08:00
Ke Bao 9ca2ae1c6c Update test skills and guide (#22189) 2026-04-06 20:30:25 +08:00
Ke Bao c4240218cb Fix ut module importing (#22176) 2026-04-06 11:53:58 +08:00
Ke Bao 896ea75820 Remove reverted test (#22058) 2026-04-03 23:51:37 +08:00
Ke Bao 47f4fd275a [CI] Fix test suite names and add suite validation (#21937) 2026-04-03 23:47:17 +08:00
Ke Bao 9f409d0749 [CI] Adjust CI server launch timeout (#22045) 2026-04-03 22:38:07 +08:00
Ke Bao b21db86e2f [CI] Fix gpu deps import in cpu test (#21950) 2026-04-03 00:06:31 +08:00
Ke Bao dbc97456ad Enable evict swa with piecewise cuda graph (#21754) 2026-03-31 20:07:16 +08:00
Ke Bao 91ab75b505 [CI] Fix ring test timeout (#21751) 2026-03-31 18:26:08 +08:00
Ke Bao acd37d8701 [CI] Fix rerun-test suite detection to skip commented registrations (#21753) 2026-03-31 18:00:53 +08:00
Ke Bao 2456889f98 Rename rerun-ut to rerun-test (#21747) 2026-03-31 17:31:55 +08:00
Ke Bao 08d3c1a134 Fix disaggregation hybrid attention ci (#21745) 2026-03-31 16:22:54 +08:00
Ke Bao 92492896a5 Fix disaggregation test bootstrap port conflict (#21271) 2026-03-24 21:14:41 -07:00
Ke Bao c1d930c028 Increase flush cache timeout in hicache CI (#21305) 2026-03-24 19:00:59 -07:00
Ke Bao 8d7453d8fb Update CODEOWNERS (#21298) 2026-03-24 17:15:25 +08:00
Ke Bao 2406ddfdb8 Add ut guide to test skills (#21130) 2026-03-22 20:55:37 +08:00
Ke Bao 05c00088e3 Cleanup broken dist-info directories in ci deps install (#20817) 2026-03-18 17:09:23 +08:00
Ke Bao 1b690836fa Add ut coverage workflow (#20804) 2026-03-18 13:25:53 +08:00
Ke Bao c42da50289 Update test guide to contribution guide (#20805) 2026-03-18 13:25:16 +08:00
Ke Bao 7f99319c56 Add evict policy ut (#20787) 2026-03-17 22:06:14 +08:00
Ke Bao 39336f5812 Precompute swa cache location (#20449) 2026-03-16 14:38:08 +08:00
Ke Bao c3483e8e97 [CI] Move existing unit tests into unit directory (#20631) 2026-03-15 23:25:18 +08:00
Ke Bao e2be31824f [CI] Add ut coverage tool (#20628) 2026-03-15 21:13:45 +08:00
Ke Bao 0cff070828 [CI] Add READMEs for unit test directory structure (#20626) 2026-03-15 20:27:05 +08:00
Ke Bao 287dc12b05 Fix hicache log metrics (#20504) 2026-03-13 16:29:58 +08:00
Ke Bao ae7c2397b9 Fix FA3 swa spec pg_size > 1 (#20369) 2026-03-12 11:42:01 +08:00
Ke Bao 4011065c65 Update CODEOWNERS (#20320) 2026-03-11 14:08:07 +08:00
Ke Bao f947bcbd89 Move stop words to args in send one (#20193) 2026-03-09 23:05:47 +08:00
Ke Bao 2e444bdced Move stop words to args in send one (#20193) 2026-03-09 23:05:32 +08:00
Ke Bao 07359efce9 Fix missing clone in hicache (#20130) 2026-03-08 23:21:18 +08:00
Ke Baoandhzh0425 5867c3fa80 Support HiCache for MambaRadixCache (#19663)
Co-authored-by: hzh0425 <hzh0425@apache.org>
2026-03-08 00:36:25 +08:00
Ke Bao 4726073865 Fix mamba2 mixer ci test (#19658) 2026-03-02 14:55:15 +08:00
Ke Bao f51e9d9ca1 Add ci test for ring model (#18829) 2026-02-14 22:20:23 +08:00
Ke Bao a0ebaa6498 Cleanup debug log for Ring model (#18793) 2026-02-13 18:36:20 +08:00
Ke Bao a9d59776cc Enhence gsm8k test (#18791) 2026-02-13 18:08:57 +08:00
Ke Bao eacab2868a Adjust mamba cache allocation (#18786) 2026-02-13 18:06:23 +08:00
Ke Bao a6c4b52ac5 Cleanup unused rerun stages (#18788) 2026-02-13 17:44:42 +08:00
Ke Bao 93ede0db19 Update ci permission (#18693) 2026-02-12 13:25:43 +08:00
Ke Bao 5d185efb78 Fix prefill stats for dllm (#18632) 2026-02-12 01:00:30 +08:00
Ke Bao d9050b4a9c Reset evict swa status when retract (#18059) 2026-02-01 17:17:37 +08:00
Ke Bao d396650bd2 Fix swa kv cache memory allocation (#18039) 2026-02-01 14:26:51 +08:00
Ke Bao 77a27e728c Add cuda graph status to prefill log (#17836) 2026-01-30 16:56:53 +08:00
Ke Bao 30ece5e1d6 Fix swa memory pool size with spec (#17630) 2026-01-25 14:10:43 +08:00
Ke Bao fb683be6eb Use attn tp group in embedding for more models (#17570) 2026-01-24 13:37:44 +08:00
Ke Bao 7ace64d1d8 Update mamba env setting (#17566) 2026-01-23 11:02:32 +08:00
Ke Bao d725487dc8 Disable swa memory for gpt-oss with spec (#17517) 2026-01-22 01:04:19 +08:00
Ke Bao a618202fc7 Tiny refine swa kv cache free (#17417) 2026-01-21 19:20:05 +08:00
Ke Bao d97066d209 Use attn_tp_group for all reduce in token embedding (#17403) 2026-01-20 23:56:21 +08:00
Ke Bao ce2d686e94 Disable swa memory for trtllm-mha backend (#17429) 2026-01-20 23:55:00 +08:00
Ke Bao 91a4cd8648 Update default attn backend to fa3 for mimo model (#17419) 2026-01-20 21:22:27 +08:00
Ke Bao ce8a6ac690 Evict swa kv cache during decoding (#17220) 2026-01-19 22:36:52 +08:00
Ke Bao f3a7c7dcd9 Move radix cache related tests (#17295) 2026-01-18 17:47:32 +08:00
Ke Bao 1fe0c82f67 Add kl test for swa radix cache (#17281) 2026-01-18 16:19:57 +08:00
Ke Bao f78201f3a9 Tiny fix comment typo (#17287) 2026-01-18 15:06:33 +08:00
Ke Bao e499258e97 Use swa radix cache and memory pool for gpt-oss model (#17261) 2026-01-18 13:47:07 +08:00
Ke Bao 7f8a58fffb Refactor prefix cache type checking (#17028) 2026-01-15 11:28:13 +08:00
Ke Bao 02722b9113 Fix evict swa for overlap scheduler and page size > 1 (#16507) 2026-01-06 22:29:40 +08:00
Ke Bao 3aa11ca722 Remove hybrid_kvcache_ratio in server args (#16399) 2026-01-06 13:13:13 +08:00
Ke Bao 76bc07a335 Move swa memory pool to a seperate file (#16347) 2026-01-04 22:39:30 +08:00
Ke Bao b328cd20bb Skip local attn init metadata for mimo swa model (#16349) 2026-01-04 22:38:36 +08:00
Ke Bao 7c1b4b1c4c Support swa allocator page size > 1 (#16296) 2026-01-03 08:49:47 +08:00
Ke Bao c483a5f45f Tiny adjust hybrid swa handling (#16292) 2026-01-02 23:11:23 +08:00
Ke Bao 698629d1f2 Add swa page table to FA3 backend forward metadata (#16267) 2026-01-02 20:00:43 +08:00
Ke Bao b6267de5ae Upgrade python version in lint ci (#16221) 2025-12-31 18:22:53 +08:00
Ke Bao b973202526 Split tp model worker init (#16165) 2025-12-30 23:36:03 +08:00
Ke Bao b3817fa93b Split model_worker init function (#16160) 2025-12-30 21:39:11 +08:00
Ke Bao 059428bd8a Tiny remove additional args in init_memory_pool (#16158) 2025-12-30 21:38:06 +08:00
Ke Bao 3de23274ee Clean up swa handling in fa3 backend (#15877) 2025-12-30 08:51:40 +08:00
Ke Bao faecd37ed4 Add Mimo-v2-flash model to ci test (#15887) 2025-12-27 14:18:08 +08:00
Ke Bao c28c536c91 Fix swa available memory check (#15867) 2025-12-26 13:19:01 +08:00
Ke Bao 7b7e357f61 Separate swa and local attention chunk cache eviction (#15820) 2025-12-26 09:34:22 +08:00
Ke Bao 2f66b0671b Fix chunk_kda_fwd missing argument (#15851) 2025-12-26 09:32:35 +08:00
Ke BaoandLiangsheng Yin cbf9f13493 Adjust server args for Mimo-v2-flash model (#15803)
Co-authored-by: Liangsheng Yin <lsyincs@gmail.com>
2025-12-25 16:04:39 +08:00
Ke Bao 8fe3e37468 Support piecewise cuda graph for dsv3 fp4 (#15531) 2025-12-21 14:50:32 +08:00
Ke Bao b399e3ac4f Support piecewise cuda graph for fused marlin moe (#15100) 2025-12-16 20:05:32 +08:00
Ke Bao 5b5571a8da Apply back moe_sum_reduce for fused_marlin_moe (#14829) 2025-12-11 00:39:41 +08:00