75 Commits
Author SHA1 Message Date
TJian 50b21a2667 [Perf] [Feat] [ROCm] Add densemha support to ROCm AITER Sparse MLA (#49263)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-10-12 15:50:28 +00:00
TJian c05d923f18 [Doc] [ROCm] Update installation documentation (#52303)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-08-17 01:56:44 -07:00
TJianandOpenAI Codex 0906123953 [ROCm] [Model] Enable TML inkling (#48841)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
2026-07-27 10:05:46 +00:00
TJian a2b1f9fc3b [ROCm] [Release] [Bugfix] Fix the per commit wheel release pipeline. (#49245)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-07-20 22:12:38 +00:00
TJian ef0aa7ca2f [ROCm] [Release] [Per-commit] Reenable per commit rocm wheel (#49044)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-07-19 13:38:04 -05:00
TJian de2a8fc042 [ROCm] [PyTorch] Move to stable abi since ROCm upgraded to torch 2.11 (#47128)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-07-02 18:34:07 +08:00
TJian 8921c4be88 [ROCm] [Performance] Optimize aiter moe for DeepSeekV4 (#46122)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-06-26 06:43:27 -07:00
TJian 302954e5f6 [ROCm] [CI] fix transcription flakiness AMD: Entrypoints Integration (API Server OpenAI - Part 1) (mi325_1) (#46823)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-06-26 21:33:35 +08:00
TJian 1bdf9810aa [ROCm] [Bugfix] Bugfix ROCm Sparse Indexer (#46222)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-06-20 13:38:42 -07:00
TJian bfe1001ab6 [Bugfix] [DSV4] [ROCm] Pin apache-tvm-ffi version to 0.1.10 (#45169)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-06-10 17:41:15 +00:00
TJian aa6fb8a329 [Bugfix] [ROCm] [Critical] fallback to regular abi for ROCm (#44648)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-06-05 15:51:17 +00:00
TJian 0ba46d4b11 [ROCm][DSV4] Enable Tilelang MHC replacing torch/triton mhc (#43679)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-05-28 07:05:28 +00:00
TJian 1806d1adfc [ROCm] [DSv4] [Perf] Support DeepSeek v4 MTP (#43385)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-05-24 18:43:08 +08:00
TJian 46f95b2ec2 [ROCm][Critical] Fix the GDN import bug (#43486)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-05-23 21:12:58 +00:00
TJian 599e75f432 [ROCm] [Bugfix] Fix DeepSeek V4 Functionality and Accuracy (#42810)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-05-17 12:18:50 -04:00
TJianandYongye Zhu a8887c208f [Bugfix] [ROCm] [DSV4] [Perf] Add aiter mhc support (#41946)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
2026-05-13 21:43:15 +08:00
TJianandCyrus Leung 445d747434 [Bugifx] Missing Renderer for fastokens mode (#41984)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
2026-05-07 23:45:14 -07:00
TJian 583e6f2226 [ROCm] [Wheel] [Bugfix] [Critical] Remove any packages installed from github from rocm.txt e.g fastsafetensors as it is incompatible with uv pip (#40461)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-04-22 00:18:07 +00:00
TJian 45232a454e [FEAT] [Perf] [Gemma4] Fused Gemma4 Routing Function Triton (#39083)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-04-19 09:57:39 +00:00
TJian 42c6bb4b75 [ROCm] [AITER] Revert AITER version to v0.1.10.post3 (#39509)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-04-10 16:25:52 +00:00
TJian 03ac6ca895 [ROCm] [DOC] Update the Documentation to include ROCm Nightly Wheel support (#38457)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-03-30 02:25:46 -07:00
TJian 58a249bc61 [ROCm] [Release] Update ROCm variant from rocm700 to rocm721 (#38413)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-03-28 06:07:03 +00:00
TJian bc9c6fbbe6 [ROCm] [Bugfix] [Release] Fix nightly rocm release pipeline (#38263)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-03-26 18:47:10 +00:00
TJianandHongxia Yang 60af7b967b [Releases] [ROCm] Enable Nightly Docker Image and Wheel Releases for ROCm (#37283)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: Hongxia Yang <hongxiay.yang@amd.com>
2026-03-26 16:32:25 +00:00
TJian da70c87e81 [CI] Fix wrong path test file, missing rlhf_async_new_apis.py (#37532)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-03-19 02:21:55 -07:00
TJianandvllmellm fb7fdc49c4 [ROCm] [CI] Add new fusion test cases that are relevant to vLLM IR Ops (#34307)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
Co-authored-by: vllmellm <vllm.ellm@embeddedllm.com>
2026-03-03 06:24:21 -08:00
TJian 5dfc5abe94 [ROCm] [Release] Change the package from aiter to amd-aiter (#35198)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-03-02 23:13:39 -08:00
TJian 5001211369 [ROCm] [CI] fix test_unrecognized_env (#34350)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-02-11 18:50:44 +00:00
TJian 785cf28fff [ROCm] [CI] Reduce Resource of two test groups (#34059)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-02-08 15:17:26 +08:00
TJian c487a8eef4 [Release] [ROCm] Remove old build step (#33316)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-28 23:35:51 -08:00
TJian f9d03599ef [Release] [CI] Optim release pipeline (#33156)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-28 22:45:42 -08:00
TJian 1ebdff412a [DOC] [ROCm] Update doc for v0.14.1 (#32998)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-25 09:13:21 +08:00
TJian 9b67338b78 [Bugfix] Suppress log on non-ROCm platform (#32703)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-20 13:38:20 -08:00
c025263ddd [Doc] [ROCm] Update ROCm getting started doc (#32580)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: Hongxia Yang <hongxia.yang@amd.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
2026-01-20 09:20:08 -08:00
TJian 41c544f78a [ROCm] [CI] [Release] Rocm wheel pipeline with sccache (#32264)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-16 02:56:18 +08:00
TJian 0346396e94 [ROCm] [Bugfix] Fix order of mori build in Dockerfile.rocm_base (#32179)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-12 15:33:21 +00:00
TJian 7a05d2dc65 [CI] [ROCm] Fix tests/entrypoints/test_grpc_server.py on ROCm (#31970)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-09 12:54:20 +08:00
TJian 72c068b8e0 [CI] [Bugfix] Fix unbounded variable in run-multi-node-test.sh (#31967)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2026-01-08 05:42:01 -08:00
TJian 578c8f51f6 [CI] [Critical] [CUDA] Fix duplicated test name (#31562)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-30 21:01:09 -08:00
TJian 022f3cea53 [ROCm] [Critical]: Remove unused variable (#31156)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-22 08:28:22 -08:00
TJian 2410132bb1 [ROCm] [Bugfix] Fix torch sdpa hallucination (#30789)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-16 15:32:43 -08:00
TJian d0fb572929 [ROCm] [AITER] [DOC] Add usage description about check functions in _aiter_ops (#30586)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-16 13:50:47 +00:00
TJian 3f1b03739a [ROCm] [Bugfix] compute_attn_mask_seqlen for qwen3 omni (#29974)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-04 08:20:24 +00:00
TJian da8e1a1bf9 [DOC] Add vLLM Bangkok Meetup info (#29561)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-27 04:42:50 +00:00
TJian 82b05b15e6 [BugFix] [FEAT] Enable fastsafetensors for ROCm platform (#28225)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-20 16:34:11 +00:00
TJian a425dc256e [Bugfix] [ROCm] [AITER]: Fix aiter block quant not compatible with torch compile dynamo (#28716)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-14 10:30:50 -08:00
TJian edb59a9470 [ROCm] [Bugfix] Fix fused_qknorm_rope_kernel rocm compatibility (#28500)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-12 05:01:14 -08:00
TJian 4673e465ff Add @tjtanaa to codeowner for ROCm and multi-modal (#28360)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-10 21:39:17 +08:00
TJian e2347dbf58 [Bugfix] [Model] Missing MRoPE function definition from KeyeForConditionalGeneration (#27895)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-11-01 13:45:23 +08:00
TJian 9c5ee91b2a [ROCm] [VL] [Bugfix] Fix vit flash attn dispatcher logic for ROCm (#26104)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-10-02 22:34:53 -07:00
TJian 6c7af8110a [Doc] Update vLLM Singapore Meetup info (#24234)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-09-04 02:58:18 -07:00
1298c67795 [FEAT] [Performance] Enable DP for ViT in Qwen2.5VL (#22742)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
2025-08-19 15:25:57 +00:00
TJian 65abe111a3 [CI] Skip Tree Attn Test in test_max_len.py to unblock CI (#22664)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-08-11 10:36:05 -07:00
TJian 42172ad18f [FEAT] [Performance] Add triton mrope to replace the torch code path (#22375)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-08-09 11:50:03 -07:00
TJian 1ee5ead5f8 [ROCm] [V1] [SpecDec] Enable Speculative Decoding on ROCm V1 Engine (#21496)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-08-07 19:13:17 -07:00
TJian 6ad6b8e115 [FEAT] Refactor ROPE into module (#22192)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-08-04 19:12:16 -07:00
TJian 26b5f7bd2a [BUG] [ROCm] Fix import bug on ROCm (#22083)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-08-01 05:25:20 -07:00
TJian e626d286f5 [FEAT] [ROCm] [AITER]: Add AITER HIP block quant kernel (#21242) 2025-07-28 05:07:06 +00:00
TJianandvllmellm c488b928a7 [ROCm] [Bugfix] [Critical]: Fix mamba compilation bug (#20883)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: vllmellm <vllm.ellm@embeddedllm.com>
2025-07-14 15:23:28 +08:00
TJian 80d38b8ac8 [V1] [ROCm] [AITER] Upgrade AITER to commit 916bf3c and bugfix APIs (#20880)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-07-13 15:19:32 +00:00
TJian 02cabff207 [V1] [ROCm] Enable EP with AITER Fused MoE (#20270)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-07-01 16:48:30 +00:00
TJian 27c065df50 [Bugfix][V1][ROCm] Fix AITER Flash Attention Backend (Fix API Break and Local Attention Logic: affecting Llama4) (#19904)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-06-26 12:42:31 +00:00
TJian 2de12be428 [ROCm] [AITER] [Bugfix] Patch for AITER commit 648764942e552a8bb5fe16026703716a81f05374 (#18990)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-06-18 22:56:31 -07:00
TJian 92540529c0 [Bugfix] [ROCm]: Remove assertion logic when using AITER fused moe in unquantizedMethod to reenable LLama4 BF16 (#18205)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-05-15 09:53:18 -07:00
TJianandGregory Shtrasberg 612c2edb4f [FEAT] [ROCm]: Add AITER CK 2 Stages MoE support (#17110)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
2025-05-14 03:03:11 -07:00
TJian 7de18d541b [BUG] [ROCm] [MLA] Fix variable name bug due to change in variable name in PR #17483 (#17961)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-05-11 09:14:30 -07:00
TJian a810b5b088 [BugFix] [ROCm]: Bugfix and handle addition case of input for rocm_aiter_rms_norm (#17857)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-05-11 04:17:11 -07:00
TJian f5a3c655b2 [FEAT] [ROCm]: Add Qwen/Qwen3-235B-A22B-FP8 TP4 triton fused moe config (#17535)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-05-01 06:37:17 -07:00
TJian 2007d4d54f [FEAT] [ROCm]: Add Qwen/Qwen3-30B-A3B-FP8 fused moe config for MI300X (#17530)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-05-01 06:03:13 -07:00
2976dc27e9 [Bug] [ROCm] Fix Llama 4 Enablement Bug on ROCm: V0 ROCmFlashAttentionImpl and Triton Fused MoE bugs (#16198)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
Co-authored-by: Hongxia Yang <hongxia.yang@amd.com>
Co-authored-by: kliuae <kuanfu.liu@embeddedllm.com>
2025-04-08 19:12:34 -07:00
TJian 4965ec42d2 [FEAT] [ROCm] Add AITER int8 scaled gemm kernel (#15433)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-03-29 03:33:56 -07:00
TJian ec870fba9a [FEAT] [ROCm]: Add AITER RMS Norm (Layer Norm) Feature (#14959)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-03-21 22:36:14 -07:00
TJian 916836bbfb [FEAT] [ROCm] [Embedding] Add encoder-only model support into ROCm Flash Attention to enable embedding models. (#14664)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-03-12 09:31:19 -07:00
TJian 848a6438ae [ROCm] Faster Custom Paged Attention kernels (#12348) 2025-03-03 09:24:45 -08:00
TJian eaa92d4437 [ROCm] [Feature] [Doc] [Dockerfile] [BugFix] Support Per-Token-Activation Per-Channel-Weight FP8 Quantization Inferencing (#12501) 2025-02-07 08:13:43 -08:00