Commit Graph

Select branches

Hide Pull Requests

br/v0.18.0

br/v0.18.0rc1

v0.11.0

Mono Color

• b6549b6e38 Add feature: priority br/v0.18.0 Jing Wang 2026-05-12 11:51:57 +00:00
• d627a45881 fix multiproc executor determine kv cache memory & update Dockerfile Jing Wang 2026-04-24 08:31:54 +00:00
• 6c097beaa5 adapt to vllm-ascend v0.18.0 Jing Wang 2026-04-21 03:05:32 +00:00
• e18643f8a4 [doc][0.18.0] v0.18.0 release note (#8383) zzzzwwjj 2026-04-30 18:44:08 +08:00
• 600bf80c6d [CI]Fix the error caused by layer_sharding in dsv32 (#8719) Nagisa125 2026-04-30 10:35:48 +08:00
• 483f7d8188 [Doc][v0.18.0] Add GLM-5.1 to models tutotials (#8778) ZYang6263 2026-04-30 10:09:55 +08:00
• 0cb4bca1ff [Doc] EPLB update the documentation (#8795) LI SHENGYONG 2026-04-29 17:15:29 +08:00
• bc5ca2c856 [0.18.0][Bugfix] Restore VLLM_MOONCAKE_ABORT_REQUEST_TIMEOUT to original value for nightly test (#8794) zhangxinyuehfad 2026-04-29 14:31:12 +08:00
• 96b90ad625 [CI] repair bug custom op ci conftest.py (#8786) ZT-AIA 2026-04-29 10:17:08 +08:00
• 73f55bd6da [Misc] add memfabric and memcache into requirements(#8748) zouyida2052 2026-04-29 09:30:15 +08:00
• 2e2aaa2fae [Doc][v0.18.0] Fix documentation formatting and improve code examples (#8701) SILONG ZENG 2026-04-28 09:01:25 +08:00
• 9a0b786f2b [bugfix][0.18.0] Fix race in non-blocking num_accepted_tokens (#8764) pppeng 2026-04-27 23:28:52 +08:00
• 9fd01a52c0 [v0.18.0][BugFix] Fix DSV3.1 W4A8 TTFT degradation (#8674) wangbj127 2026-04-27 23:27:34 +08:00
• 0cc76860d5 [CI ][Misc] Add timeout check for custom op CI and optimize test parameters (#8755) ZT-AIA 2026-04-27 21:48:54 +08:00
• 8a671a109c [CI][Cherry-pick] Relax TTFT benefits threshold from 0.4 to 0.5 to account for DP load imbalance (#8684) hucong 2026-04-27 16:38:07 +08:00
• 2cee0c32e5 [CI] Repair custom op nightly (#8707) ZT-AIA 2026-04-25 19:05:33 +08:00
• 963b82f42d [Doc]Update Qwen3-Omni-30B-A3B-Thinking.md (#8669) tanhaoan333 2026-04-25 14:25:46 +08:00
• e17006077a fix multiproc executor determine kv cache memory & update Dockerfile br/v0.18.0rc1 starkwj 2026-04-24 08:31:54 +00:00
• 81d0a37bf5 [CI] repair ci custom op (#8571) ZT-AIA 2026-04-24 17:06:25 +08:00
• bd3774d601 [Doc][Misc] Improve documentation quality by revising specific content. (#8603) Lucky1 2026-04-24 15:40:41 +08:00
• 97dbcaf919 [BugFix][310P][v0.18.0] Use CPU generator cache for sampling (#8624) csoulnd 2026-04-24 09:34:14 +08:00
• 00ddacf4e7 [Doc][0.18.0] Add the 310p guide (#8639) Shaoxu Cheng 2026-04-23 23:50:40 +08:00
• 571edc58fa [Doc]Update DeepSeekOCR2.md for releases/v0.18.0 (#8604) Wangbei25 2026-04-23 23:48:03 +08:00
• 696dcc9265 [Bugfix][0.18.0] fix kernels in sample when mask is not static or draft_token_id is invalid (#8531) pppeng 2026-04-23 23:04:19 +08:00
• 45f75b4178 [Doc][v0.18.0]Fix minimax2.5 readme (#8638) aipaes 2026-04-23 22:56:24 +08:00
• c47371c1af [BugFix]Backport validate pd mode feature gates no fused mc2 v0.18.0 clean (#8583) wangxiaoteng888 2026-04-23 19:44:07 +08:00
• 1fc7bc056d [0.18.0][Doc] Add NPU soft partitioning + cudagraph.piecewise limitation (#8595) zzzzwwjj 2026-04-23 19:09:55 +08:00
• 5c048a9b71 [Doc][releases/v0.18.0] fix documentation error or non-standard description (#8626) linfeng-yuan 2026-04-23 18:55:44 +08:00
• 786eaf8b07 [Doc][Misc] Improve readability and fix typos in documentation (#8633) sunshine202600 2026-04-23 18:45:17 +08:00
• d81101acdd [releases/v0.18.0][Platform][BugFix] Guard forced tool choice with empty content (#8400) jack 2026-04-23 16:46:10 +08:00
• ff76c6780e [releases/v0.18.0][Doc][Misc] Modifying Configuration Parameters (#8618) herizhen 2026-04-23 16:23:31 +08:00
• ce92be29d2 [Doc] Clarify irqbalance service management (#8614) Frank Chen 2026-04-23 16:10:07 +08:00
• a4ba82e138 [BugFix] Require kv producer for layer sharding (#8563) Frank Chen 2026-04-23 16:06:53 +08:00
• 4a254ba59a [Doc] [v0.18.0]Fix glm4.7 readme v18 (#8460) aipaes 2026-04-23 14:42:28 +08:00
• 58c87bd15b [BugFix][0.18.0] Remove unused layers assignment in mooncake connector (#8602) liziyu 2026-04-23 14:08:53 +08:00
• 30d08ced2d [Doc][0.18.0] Fix kv pool CLI flag typo and formatting (#8608) pz1116 2026-04-23 12:07:47 +08:00
• 0c458aa6dc [v0.18.0][Doc] Translated Doc files 2026-04-22 (#8565) vllm-ascend-ci 2026-04-23 11:06:05 +08:00
• 9e31e4f234 [Doc]change format (#8592) Yang Yuxi 2026-04-23 10:46:09 +08:00
• 4020b3df60 [BugFix] fix tl.extract_slice and tl.insert_slice. (#8567) wangx700 2026-04-23 09:50:29 +08:00
• c3b1d409a9 [BugFix] [P/D] [CherryPick] 8540 In scenarios where TP is not equal, the KV cache at the MTP layer is not handled. (#8541) liziyu 2026-04-23 09:16:37 +08:00
• fcf4d477a7 [BugFix][0.18.0]dispatch_ffn_combine kernal rollback combine 、unpermute part and scale part (#8534) LQLlulu 2026-04-22 23:27:02 +08:00
• 69a57bc9ec [Doc][0.18.0]Fix typo in KV Cache Pool developr guide (#8575) pz1116 2026-04-22 17:59:16 +08:00
• c335710b82 [Misc] Bump mooncake version to v0.3.9 (#8445) Li Wang 2026-04-22 15:41:06 +08:00
• 9e15ce7074 [Bugfix] [P/D] Fix connector with pcp dcp (#8538) liziyu 2026-04-21 22:23:21 +08:00
• 2cb9f76a0f [CI] Ds32 ep aime2025 (#8496) starmountain1997 2026-04-21 19:49:06 +08:00
• 862bd8143c [Releases/v0.18.0][Doc][Misc] Added version description in the writing template (#8451) herizhen 2026-04-21 16:33:43 +08:00
• 7850264324 [v0.18.0][BugFix] PIECEWISE mode also requires synchronization (#8469) 1kzk 2026-04-21 16:22:32 +08:00
• b717dc17a3 [v0.18.0][Test][Misc] Update CI for GLM-5 configuration on vllm-ascend/releases/v0.18.0 branch (#8322) yangjiuhua 2026-04-21 14:10:11 +08:00
• e4d898b245 adapt to vllm-ascend v0.18.0rc1 starkwj 2026-04-21 03:05:32 +00:00
• 36a0470de1 [Doc] Upgrade env VLLM_ASCEND_ENABLE_FUSED_MC2 used in nightly test and tutorials (#8441) Li Wang 2026-04-20 22:39:23 +08:00
• 3db5048d74 [CI]repair ci for custom op (#8455) ZT-AIA 2026-04-20 17:51:37 +08:00
• dd7e08c6db [Performance] Use forward_native for Conv3dLayer and add UT (#8375) zyz111222 2026-04-20 17:20:40 +08:00
• c124e8df07 Revert "[BugFix] dispatch_ffn_combine kernal rollback combinev2 part … (#8439) LQLlulu 2026-04-20 14:26:55 +08:00
• 7a706fb197 [v0.18.0][CI] fix report_template.md (#8429) zhangxinyuehfad 2026-04-20 11:13:14 +08:00
• e6ba5a88f7 [v0.18.0][BugFix] Fix Qwen3.5 MoE FC1 error under high concurrency when dp>1 (#8395) wangbj127 2026-04-20 10:26:19 +08:00
• b992b11545 [BugFix] dispatch_ffn_combine kernal rollback combinev2 part (#8405) LQLlulu 2026-04-18 22:45:08 +08:00
• 6bdc72949b Revert "[v0.18.0][BugFix] Fix dimension mismatch error when SP padding causes num_tokens_padded != num_tokens_unpadded" (#8413) wangbj127 2026-04-18 20:43:42 +08:00
• 363febb6cb [BugFix][v0.18.0] Gate recompute/balance/fused_mc2 by PD mode (#8374) wangxiaoteng888 2026-04-18 18:06:42 +08:00
• c995a959e6 [BugFix] fix hang in async scheduling while open ENPU (#8354) 1kzk 2026-04-18 00:07:15 +08:00
• f81f9a3c89 [Doc] Add Qwen3.5 fused MC2 known issue for 0.18.0 release (#8378) shaopeng-666 2026-04-17 22:54:21 +08:00
• f2956ce944 [v0.18.0][BugFix] Fix dimension mismatch error when SP padding causes num_tokens_padded != num_tokens_unpadded (#8133) wangbj127 2026-04-17 22:50:22 +08:00
• 0954fd0912 [BugFix][0.18.0] Fix quant_bias missing in w8a8_static when flashcomm1 is enabled for GLM-5 (#8304) aipaes 2026-04-17 22:46:36 +08:00
• b72ade9acd [0.18.0][BugFix] Update capture sizes after rounding operations (#8380) Zetong Li 2026-04-17 22:46:16 +08:00
• 76cc2204bd [Doc] Sensitive word modification (#8303) herizhen 2026-04-17 16:30:00 +08:00
• 9c1d58f4d2 [v0.18.0][Doc] Translated Doc files 2026-04-15 (#8309) vllm-ascend-ci 2026-04-17 16:29:30 +08:00
• ceb1e49661 [BugFix][v0.18.0] fix remote KV waiting promotion in balance scheduler (#8280) pz1116 2026-04-17 10:06:36 +08:00
• f85144cc57 [BugFix][DSv32] Fix DSA-CP PD role gating for deepseek v3.2 (v0.18.0) (#8291) Frank Chen 2026-04-17 10:05:40 +08:00
• 1dd1de8153 [Doc][Misc] Improve readability and fix typos in documentation (#8340) sunshine202600 2026-04-17 08:54:38 +08:00
• 8952fddc7e [BugFix][310p][Cherry-pick] Handle null quantization config in ShardedStateLoader310&[Feature][310P] Support W8A8 dynamic linear method (#8296) csoulnd 2026-04-16 16:53:39 +08:00
• 52f0f9b5e4 [0.18.0][BugFix]: order acl graph updates before model forward for ENPU (#8317) 1kzk 2026-04-16 16:26:59 +08:00
• 2ac8bfb4cb [BugFix] Enforce C locale for CPU binding subprocess parsing (#8261) Frank Chen 2026-04-16 16:17:10 +08:00
• 8bc72a807a [BugFix][v0.18.0] require piecewise cudagraph for layerwise AscendSto… (#8282) pz1116 2026-04-16 10:40:14 +08:00
• 028b8cabc4 [BugFix][Platform] Fix extra function name in final chunk of streaming tool calls (#8178) chenweiqiang11 2026-04-15 17:50:10 +08:00
• 808d00406f [v0.18.0][CI]Add rank0 process count check for DeepSeek-R1-W8A8-HBM test (#8072) zhangxinyuehfad 2026-04-15 17:16:27 +08:00
• 95726d20eb [Doc][Misc] Correcting the document and uploading the model deployment template (#8287) herizhen 2026-04-15 16:03:11 +08:00
• 147b589f62 [v0.18.0][Doc] Translated Doc files 2026-04-14 (#8257) vllm-ascend-ci 2026-04-15 15:27:09 +08:00
• b6aa5bbdbf [0.18.0][BugFix] Add PrefillNoCache state in mla _forward_decode for short prompt (#8264) Zetong Li 2026-04-15 09:23:52 +08:00
• 70713c3fd4 [cherry-pick][BugFix] Improve max_cudagraph_capture_size validation (#8252) Qiu 2026-04-14 22:00:10 +08:00
• 13c7392416 [BugFix] fix dsv3.1 service failed to start (#8207) Yaphets24 2026-04-14 17:52:55 +08:00
• d94b1dc2d0 [v0.18.0][BugFix] Fix Qwen3.5 MoE flash comm v1 shared expert shape error of mtp layer on A2 (#8004) wangbj127 2026-04-13 17:36:09 +08:00
• 39c071a0f5 [BugFix][P/D][0.18.0]Add a retry mechanism to prevent packet loss (#8167) wangxiaoteng888 2026-04-12 23:30:45 +08:00
• 4adc6a68f5 [BugFix][P/D][0.18.0]bugfix short squence has no respone (#8142) wangxiaoteng888 2026-04-12 23:25:01 +08:00
• 31186a3a9d [BugFix] Add async communication check for capturing mode (#8149) Frank Chen 2026-04-12 21:52:54 +08:00
• c1f323ee46 [Doc] Add new intro to MiniMax-M2.5/M2.7 (#8169) SparrowMu 2026-04-12 21:45:07 +08:00
• 9e8da00f95 [V0.18.0][Doc] add preemption in FAQs (#8136) Angazenn 2026-04-10 17:36:45 +08:00
• 99cea6c1b5 [CI] Fix the nightly pip binary install doc test fail. (#8129) Nengjun Ma 2026-04-10 17:34:18 +08:00
• bd9927d5a9 [releases/v0.18.0][Build][BugFix] support ascend950 npu-smi info interface changes and make SOC_VERSION actually take effect (#8061) linfeng-yuan 2026-04-10 16:44:38 +08:00
• 34386c8896 [v0.18.0][CI] Fix and simplify the CI for Qwen3 32B (#8093) ZYang6263 2026-04-10 14:22:24 +08:00
• 531d0e6fff [v0.18.0][BugFix][KV Pool]Fix the conflict between pooling scenarios … (#8101) DreamerLeader 2026-04-09 21:55:56 +08:00
• 054fde7b72 [0.18.0][BugFix] Fix attention state of short prompt for correct forwarding (#8088) Zetong Li 2026-04-09 21:21:24 +08:00
• f668ff9ef0 [v0.18.0][BugFix]Revert the code: Replace npu_ring_mla wit FIA with MLA prefill. (#7961) weijinqian0 2026-04-09 17:00:25 +08:00
• 7c9aa498d6 [releases/v0.18.0][BugFix] Restore global_bs=0 and mc2_mask for uniform-token dispatching and support inter-node roce hierarchical MC2 communication (#8040) linfeng-yuan 2026-04-09 16:51:17 +08:00
• 82e17f693a [BugFix][0.18.0][310p] fix post-sampling not working in graph mode on 310p (#8077) Shaoxu Cheng 2026-04-09 16:31:38 +08:00
• 0d1424d81a [Doc][Misc] Comprehensive documentation cleanup and grammatical fixes (#8073) herizhen 2026-04-09 15:37:57 +08:00
• c40a387f63 [bugfix]fix extra npu context in device 0 (#8041) zouyida2052 2026-04-08 23:35:52 +08:00
• 4a628f1042 [UT][v0.18.0] Fix APC nightly UT and TTFT ratio (cherry-pick #7468) (#8053) hucong 2026-04-08 21:08:26 +08:00
• 044d4c3974 [v0.18.0]feat(quant): add C8 INT8 KV cache support for GQA attention models (#7474) (#8007) Mengqing Cao 2026-04-08 10:51:58 +08:00
• fbd5d0fd55 [Doc][Misc][v0.18.0] Updated the document configuration for DeepSeek-V3.2 (#7970) Nagisa125 2026-04-07 16:17:28 +08:00
• 6c19270498 [BugFix] fix qwen3-next compilation error (#7977) cvSoldier 2026-04-03 20:03:39 +08:00
• 81c6f51a45 【CI】add nightly cases: MiniMax-M2.5-W8A8 Qwen3.5-27B-w8a8 Qwen3.5-397B-A1… (#7968) guxin108 2026-04-03 17:50:59 +08:00