 yhcheong (vllmellm)
|
273e099693
|
[Bugfix][ROCm] Select page-aligned kernel blocks for pooled indexers so the block table addresses storage pages (GLM-5.3-Flash) (#59412)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-10-08 14:44:58 +00:00 |
|
|
yhcheong (vllmellm)
|
16d4ac4ba3
|
[ROCm][MoE] Support MiMo-V2.6 MXFP4 on gfx942 (#58262)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-09-30 17:23:00 +08:00 |
|
|
yhcheong (vllmellm)
|
77e52645e9
|
[Bugfix][Model] MiMo: keep fused fp8 qkv_proj pairing state across weight-loading calls (#58142)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-09-29 09:37:20 +00:00 |
|
|
vllmellm
|
211e252d0b
|
[Bugfix][Model] Fix MiMo-V2.5 fused fp8 qkv_proj sharding (pre-shard count is num_key_value_heads; MTP path too) (#57508)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-09-19 06:05:00 -07:00 |
|
|
vllmellmandTJian
|
3affd35f31
|
[Fix][ROCm] MXFP4 MoE round-up inflates TP-sharded expert weights on CDNA3, starving KV cache (#56359)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-09-17 22:52:12 +08:00 |
|
|
vllmellm
|
f3c1638927
|
[ROCm] Enable V2 model runner for Kimi-K3 on ROCm (#51653)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-08-13 14:34:53 +00:00 |
|
|
vllmellmandtjtanaa
|
436be94e13
|
[ROCm][Bugfix] Use TCP store when AITER custom all-reduce is enabled (#51635)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2026-08-10 11:07:14 +00:00 |
|
|
|
c8db00b16c
|
Fix GPTQ quantized Qwen3.5 MTP weight loading with spec decode (#48816)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
Co-authored-by: noobHappylife <64898326+noobHappylife@users.noreply.github.com>
|
2026-07-23 06:59:48 -07:00 |
|
|
vllmellm
|
d537122398
|
[ROCm][Bugfix]: Fallback GFX942 sparse MLA ops to Triton (#45782)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-06-17 11:41:29 +00:00 |
|
|
vllmellm
|
25c53d1293
|
[ROCm][Doc] Add installation notes about python version requirement (#45671)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-06-15 17:22:55 +08:00 |
|
|
vllmellm
|
7f95e66a11
|
[ROCm][Bugfix]: dynamically align BLOCK_DMODEL with Lv in MLA decode kernel (#41119)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-05-11 11:14:19 +08:00 |
|
|
|
529c671e80
|
[ROCm][FEAT] AITER Fused Allreduce + RMSNorm (#37646)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: Rita Brugarolas Brufau <rita.brugarolasbrufau@amd.com>
Signed-off-by: junkang1991 <junkangchow@gmail.com>
Co-authored-by: Rita Brugarolas <Rita.BrugarolasBrufau@amd.com>
Co-authored-by: junkang1991 <junkangchow@gmail.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-05-01 23:07:18 +08:00 |
|
|
vllmellm
|
5f7fab881a
|
[ROCm][FEAT] Integrate aiter gemm w8a8 ptpc (#33773)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-04-16 09:55:29 +08:00 |
|
|
vllmellm
|
551b3fb39f
|
[ROCm] Enable VLLM triton FP8 moe for gfx1201, tuned for Qwen3-30B-A3B-FP8 tp=2 and Qwen/Qwen3.5-35B-A3B-FP8 tp=2 (#38086)
Signed-off-by: big-yellow-duck <jeffaw99@hotmail.com>
Co-authored-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-04-02 08:13:42 +00:00 |
|
|
vllmellm
|
42e9547976
|
[ROCm][Test] Fix ROCM_AITER_UNIFIED_ATTN attn+quant fusion test (#37640)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-03-25 05:06:15 +00:00 |
|
|
vllmellm
|
ee8a29511f
|
[Bugfix] Fix compressed-tensors quantization failure for DeepSeek-R1 on MI300x (#36247)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-03-07 09:26:59 +00:00 |
|
|
|
aaa2efbe98
|
[DOC] [ROCm] Update docker deployment doc (#33971)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-02-06 10:05:35 -08:00 |
|
|
vllmellm
|
174f16700b
|
[Doc] [ROCm] Update Documentation to reflect v0.15.0 release (#33388)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-30 19:06:08 +08:00 |
|
|
vllmellm
|
148117ea2e
|
[Refactor] Make FP8 Linear Ops use kernel abstraction (#27814)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-20 14:48:20 +08:00 |
|
|
vllmellm
|
e27078ea80
|
[Bugfix][ROCm][performance] Resolve the performance regression issue of the Qwen3-Next-80B-A3B-Thinking under rocm_atten (#32336)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-14 19:32:48 +00:00 |
|
|
vllmellm
|
1a19e9cd87
|
[Bugfix][ROCm]Fix Qwen3-Next-80B-A3B-Thinking inference and optimize non-standard block size (544) support under rocm_atten (#31380)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-09 19:28:02 +08:00 |
|
|
vllmellm
|
41cfa50632
|
[ROCm][AITER] fix wrong argument passed to AITER flash_attn_varlen_func (#31880)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-07 11:25:03 +00:00 |
|
|
vllmellm
|
6409004b26
|
[ROCm][AITER] bugfix accuracy regression in ROCM_AITER_TRITON_MLA backend (#31816)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-07 05:04:53 +00:00 |
|
|
vllmellm
|
e9717801bd
|
[Bugfix][ROCm] Fix Unsupported attention metadata type for speculative decoding in eagle.py (#31714)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-01-06 07:53:22 +00:00 |
|
|
vllmellmandTJian
|
f32cfd7d97
|
[ROCm][FEAT] Support AITER RMSNorm quantization fusion pass (#26575)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2025-12-23 02:07:54 -08:00 |
|
|
vllmellm
|
96bf50a2c0
|
[ROCm] Serving Fails on Radeon Due to AITER Dtype Import (#30952)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-12-18 11:47:46 +00:00 |
|
|
vllmellm
|
ee14644ba9
|
[ROCm] Aiter Quant Kernels (#25552)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-12-09 14:27:37 +00:00 |
|
|
vllmellm
|
64deead719
|
[Bugfix] [ROCm] [UX]: revert Flex attention backend (#29371)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-11-25 06:56:06 +00:00 |
|
|
vllmellm
|
e48b2e6848
|
[Bugfix] [ROCm] [UX] Reorganize ROCm Backend Selection Logic (#26980)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-11-24 15:24:49 +00:00 |
|
|
vllmellm
|
0af3d4f0df
|
[FEAT] [AITER] [ROCm] integrate aiter sampling ops (#26084)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-11-18 17:28:34 +00:00 |
|
|
vllmellm
|
d8140b9833
|
[ROCM] Fix ROCm warnings, environment flag access, and GEMM kernel naming for consistency in _aiter_ops.py (#28464)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-11-12 21:46:57 +00:00 |
|
|
vllmellmandLuka Govedič
|
f080a83511
|
[RFC][ROCm][AITER] Keep all AITER kernels in _aiter_ops class like _custom_ops and _ipex_ops (#24490)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
|
2025-11-10 08:20:53 -08:00 |
|
|
vllmellm
|
b13a447546
|
[Bugfix][ROCm] Fix ViT rotary embeddings for torch.compile compatibility on ROCm (#27748)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-11-03 17:12:19 -08:00 |
|
|
vllmellm
|
5b3c35a68e
|
[ROCm] [Doc] Update ROCm installation docs (#27327)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-10-28 13:00:50 +08:00 |
|
|
vllmellm
|
265ecb05fb
|
[DOC] [ROCm] Add ROCm quickstart guide (#26505)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-10-22 03:10:48 +00:00 |
|
|
vllmellm
|
e33ee23ee3
|
[Bugfix] [AITER] [ROCm] Fix Quark MoE Quant Config and AITER Fused MoE quant type logic (#27029)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-10-17 12:51:10 -06:00 |
|
|
vllmellm
|
4821ac1b4d
|
[CI] [ROCm] Automate CC list for ROCm related issue (#26753)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-10-14 13:57:26 +08:00 |
|
|
vllmellm
|
5e4a8223c6
|
[Qwen][ROCm] Flash Attention Rotary Embeddings (#24642)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-10-02 08:26:08 -07:00 |
|
|
vllmellm
|
0d9fe260dd
|
[docs] Benchmark Serving Incorrect Arg (#25474)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-09-23 06:05:11 -07:00 |
|
|
vllmellm
|
8c54610265
|
[Bug] [Spec Dec]: Fix kv_cache dtype mismatch for Eagle3 drafter on FP8 target (#24505)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-09-16 04:45:38 +00:00 |
|
|
vllmellm
|
7c195d43da
|
[ROCm][Bugfix] Fix Aiter RMSNorm (#23412)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-09-10 21:08:03 +08:00 |
|
|
vllmellm
|
72a69132dc
|
[CI] Add aiter to matching list of issue auto labeller for rocm tag (#23942)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-08-29 15:29:21 +00:00 |
|
|
vllmellmandCyrus Leung
|
44ac25eae2
|
[CI] [Doc]: Add GH Action for auto labeling issues with rocm tag (#20988)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
|
2025-08-26 16:20:13 +00:00 |
|
|
vllmellm
|
9c97a1c349
|
[ROCm][AITER] Support AITER Rope ops in RotaryEmbedding Module. (#22521)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-08-10 22:52:34 -07:00 |
|
|
vllmellmandkf
|
cbc8457b26
|
[Model] Switch to Fused RMS norm in Qwen2.5_VL model. (#22184)
Signed-off-by: kf <kuanfu.liu@embeddedllm.com>
Signed-off-by: tjtanaavllm <tunjian.tan@amd.com>
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: kf <kuanfu.liu@embeddedllm.com>
|
2025-08-06 23:05:24 -07:00 |
|
|
vllmellmandtjtanaavllm
|
d3a6f2120b
|
[FEAT][ROCm] Enable running Flash Attention as ViT attn backend for Qwen-VL models on ROCm platform. (#22069)
Signed-off-by: tjtanaavllm <tunjian.tan@amd.com>
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: tjtanaavllm <tunjian.tan@amd.com>
|
2025-08-01 23:53:18 -07:00 |
|
|
vllmellmandkf
|
ee2eb6ecd8
|
[Model] Qwen2.5 VL SiLU-and-Mul (#22066)
Signed-off-by: kf <kuanfu.liu@embeddedllm.com>
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: kf <kuanfu.liu@embeddedllm.com>
|
2025-08-01 19:34:37 -07:00 |
|
|
vllmellm
|
a1aafc827a
|
[ROCm][FEAT] Enable Full Graph Mode in AITER MLA V1 Attn Backend (Decode Phase only) (#20254)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-07-02 16:25:46 +00:00 |
|
|
vllmellm
|
18093084be
|
[Misc] Remove unnecessary fallback to prefill-decode attention (#19138)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-06-05 16:08:26 +08:00 |
|
|
vllmellm
|
0f5e0d567e
|
[FEAT][ROCm] Add AITER grouped topk for DeepSeekV2 (#18825)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-05-31 03:39:31 -07:00 |
|
|
vllmellm
|
77b6e74fe2
|
[ROCm] Remove unnecessary assertion of max_model_len in ROCM_AITER_MLA attention backend. (#18938)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-05-29 22:33:17 -07:00 |
|
|
vllmellm
|
d260f799a9
|
[FEAT] [ROCm] Upgrade AITER Fused MoE kernels. (#18271)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-05-26 23:14:07 -07:00 |
|
|
vllmellmandLuka Govedič
|
94d8ec8d2b
|
[FEAT][ROCm] Upgrade AITER MLA v1 backend (#18338)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
|
2025-05-21 10:34:28 -07:00 |
|
|
vllmellmandtjtanaa
|
2d912fb66f
|
[FEAT] [ROCm] [V1]: Add AITER biased group topk for DeepSeekV3 (#17955)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2025-05-13 22:03:47 -07:00 |
|
|
vllmellmandtjtanaa
|
40de1ef455
|
[FEAT] [ROCm]: Add AITER Block-Scaled GEMM Feature (#14968)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2025-05-13 19:08:20 -07:00 |
|
|
vllmellm
|
217db4baa6
|
[Bugfix][ROCm] Fix AITER MLA V1 (#17880)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-05-09 08:38:21 +00:00 |
|
|
|
3c9396a64f
|
[FEAT][ROCm]: Support AITER MLA on V1 Engine (#17523)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: qli88 <qiang.li2@amd.com>
Co-authored-by: Hongxia Yang <62075498+hongxiayang@users.noreply.github.com>
|
2025-05-09 10:42:05 +08:00 |
|
|
vllmellmandtjtanaa
|
eef364723c
|
[FEAT] [ROCm]: AITER Fused MOE V1 Support (#16752)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2025-04-25 11:06:50 +08:00 |
|
|
vllmellm
|
047797ef90
|
[Bugfix] Triton FA function takes no keyword arguments (#16902)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-04-22 21:35:24 -07:00 |
|
|
vllmellm
|
7e081ba7ca
|
[BugFix] Revert ROCm Custom Paged Attention Env Flag Check (#17022)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2025-04-22 19:17:48 -07:00 |
|
|
vllmellmandqli88
|
30bc3e0f66
|
[FEAT][ROCm]: Support AITER MLA (#15893)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: qli88 <qiang.li2@amd.com>
|
2025-04-22 09:31:13 -07:00 |
|
|
vllmellmandtjtanaa
|
0e237f0035
|
[FEAT][ROCm] Integrate Paged Attention Kernel from AITER (#15001)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2025-04-22 02:46:28 -07:00 |
|
|
vllmellmandtjtanaa
|
5ebf66748b
|
[FEAT][ROCm] Integrate Fused MoE Kernels from AITER (#14967)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
|
2025-03-26 16:30:30 +08:00 |
|
|
|
2bb0e1a799
|
[Bugfix][ROCm] running new process using spawn method for rocm in tests. (#14810)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
|
2025-03-17 11:33:35 +00:00 |
|