 HAI
|
3cafb02e5b
|
AMD/ROCm: update daily image repo to lmsysorg/sglang-rocm (#20720)
|
2026-03-16 20:04:52 -07:00 |
|
|
HAI
|
a0f3361023
|
Update (#19351)
|
2026-02-25 11:37:07 -08:00 |
|
|
HAI
|
6a999dbdf8
|
[AMD] ENV flags tuning and cleanup (#19176)
|
2026-02-22 22:40:00 -08:00 |
|
|
HAIandLiangsheng Yin
|
b2573fe426
|
Upd: CODEOWNERS (#19055)
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
|
2026-02-21 07:51:53 +08:00 |
|
|
HAI
|
0215d47007
|
[AMD] ROCm7.2: Add /sgl-workspace/aiter to PYTHONPATH (#18972)
|
2026-02-18 02:21:39 -08:00 |
|
|
HAI
|
934b36693c
|
Reasoning models fix docs (#18963)
|
2026-02-17 23:05:55 -08:00 |
|
|
HAI
|
8bb1037796
|
ROCm use rotary_embedding from sgl-kernel (#18920)
|
2026-02-17 03:00:37 -08:00 |
|
|
HAI
|
b158f5d4a2
|
Revert "[AMD] Fix RotaryEmbedding crash on AMD/ROCm (regression from #17934)" (#18922)
|
2026-02-17 01:07:50 -08:00 |
|
|
HAI
|
f4417475b8
|
Build ROCm7.2 Image with latest AITER v0.1.10.post3 (#18741)
|
2026-02-12 14:30:13 -08:00 |
|
|
HAI
|
750ad0d290
|
[AMD] enable MoRI to release and nightly builds (#18101)
|
2026-02-02 00:28:30 -08:00 |
|
|
HAI
|
65d376b491
|
aiter update to v0.1.6.post1 (#12004)
|
2025-10-22 23:53:05 -07:00 |
|
|
HAI
|
d500eb9173
|
aiter v0.1.5.post2 (#10563)
|
2025-09-17 22:10:45 -07:00 |
|
|
HAI
|
44426e54be
|
Update REVIEWERS (#9063)
|
2025-08-11 11:04:39 -07:00 |
|
|
|
b819381fec
|
AITER backend extension and workload optimizations (#6838)
Co-authored-by: wunhuang <wunhuang@amd.com>
Co-authored-by: Hubert Lu <Hubert.Lu@amd.com>
|
2025-06-05 23:00:18 -07:00 |
|
|
HAI
|
183d9f969c
|
DeepSeek: enable none block-quant FP8 quantizations (#6638)
|
2025-05-27 09:06:40 -07:00 |
|
|
HAI
|
5c0b38f369
|
aiter attention-backend (default enabled on AMD/ROCm) (#6381)
|
2025-05-20 22:52:41 -07:00 |
|
|
HAI
|
6317c5c61f
|
Address performance regression: disable multiple streams on ROCm (#6412)
|
2025-05-19 21:16:20 -07:00 |
|
|
HAI
|
d364b9b0f2
|
ROCm: update AITER (#5816)
|
2025-04-28 11:01:20 -07:00 |
|
|
HAI
|
b0feda090c
|
Revert "Support aiter RMSNorm in AMD" (#5646)
|
2025-04-22 15:20:24 -07:00 |
|
|
HAI
|
8879944800
|
ROCm/AITER CK_MoE: update 2-stage kernels & support both Activations (#5228)
|
2025-04-10 18:19:57 -07:00 |
|
|
HAI
|
d050df368c
|
ROCm sgl-kernel: compatible to later torch (#5167)
|
2025-04-10 09:18:36 -07:00 |
|
|
HAIandLianmin Zheng
|
819924748a
|
Fix refactor error - fp8.py (#5106)
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
|
2025-04-07 00:34:08 -07:00 |
|
|
HAIandlinsun12
|
0beea4503f
|
ROCm: Flex Attention Enablement with custom backends (#4178)
Co-authored-by: linsun12 <linsun12@amd.com>
|
2025-03-07 04:38:53 -08:00 |
|
|
HAI
|
13bc39c5d6
|
ROCm: enable trillion-parameter MoE models with INT4-FP8 single node (#4152)
|
2025-03-06 15:33:02 -08:00 |
|
|
HAI
|
71ab0dabe0
|
Fix the moe padding conditional logic (#4081)
|
2025-03-05 10:56:51 -08:00 |
|
|
HAI
|
51d25405a7
|
ROCm: update aiter and its usage to fused moe (bloat16, fp8, fp8 block-quant) (#4053)
|
2025-03-04 03:00:46 -08:00 |
|
|
HAI
|
5c54ef0352
|
AMD/ROCm: update AITER repo to ROCm/aiter (#3747)
|
2025-02-21 00:18:08 -08:00 |
|
|
HAI
|
6252ade985
|
revert BLOCK and num_warps on HIP (#3722)
|
2025-02-20 23:30:18 +08:00 |
|
|
HAI
|
d973c78e79
|
ROCm docker: triton update (#3584)
|
2025-02-14 10:26:32 -08:00 |
|
|
HAI
|
d81ac4434e
|
MI30x: More graph captures for larger batch sizes and concurrencies (#3420)
|
2025-02-12 03:04:38 +08:00 |
|
|
HAI
|
2c1a695ff1
|
ROCm: sgl-kernel enablement starting with sgl_moe_align_block (#3287)
|
2025-02-04 21:44:44 +08:00 |
|
|
HAI
|
566d61d90f
|
ROCm: bump 6.3.0 (#3259)
|
2025-02-03 04:13:40 +08:00 |
|
|
HAI
|
17dbf976c5
|
update ENV to ROCm dockers (#3248)
|
2025-02-01 17:27:43 +08:00 |
|
|
HAI
|
6d08ce2aa9
|
Use Optional with None default (#2770)
|
2025-01-07 01:35:08 -08:00 |
|
|
HAI
|
c5210dfa38
|
AMD DeepSeek_V3 FP8 Numerical fix (#2667)
|
2024-12-30 21:31:12 +08:00 |
|
|
HAI
|
e6f523b5f2
|
fix typo in python/sglang/srt/layers/quantization/fp8.py (#2655)
|
2024-12-29 23:45:02 -08:00 |
|
|
HAI
|
30828e7192
|
AMD: set weights and scaling numbers properly for block FP8 (#2637)
|
2024-12-29 03:23:39 -08:00 |
|
|
HAI
|
7722c11c1d
|
Regression fix to AMD/ROCm from recent change (#2606)
|
2024-12-26 20:22:14 -08:00 |
|
|
HAI
|
95f93f493a
|
Fp8 MoE optimizations on AMD (#2388)
|
2024-12-07 21:18:26 +08:00 |
|
|
HAI
|
b2986d7aa5
|
Adding SGLang FP8 Utils (#2348)
|
2024-12-04 03:01:33 -08:00 |
|
|
HAI
|
0639bf15d1
|
ROCm Container: set SGLANG_SET_CPU_AFFINITY=1 (#2328)
|
2024-12-02 23:20:33 -08:00 |
|
|
HAI
|
69e2d4fb66
|
Relax to include more AMD GPUs (#2319)
|
2024-12-02 19:05:58 -08:00 |
|
|
HAI
|
c54bda300a
|
Use rocminfo instead of rocm-smi for more OS/WSL support (#2310)
|
2024-12-02 00:15:45 -08:00 |
|
|
HAI
|
b79fffdcb5
|
Update Install Method 2. From source (#2232)
|
2024-11-27 22:46:55 -08:00 |
|
|
HAI
|
cd51758fad
|
Rename tuned MI300X config files for fused_moe_triton (#2228)
|
2024-11-27 21:18:51 -08:00 |
|
|
HAI
|
10189d08dd
|
[Performance]: Process affinity to CPU cores with multiple sockets support (#2171)
|
2024-11-25 14:57:32 -08:00 |
|
|
HAI
|
f35cb46cc3
|
ROCm: Fix MoE padding for none FP8 cases (#2111)
|
2024-11-21 12:23:21 -08:00 |
|
|
HAI
|
e57c3e12b8
|
Use native fp8 format on MI300X (#2094)
|
2024-11-19 14:06:29 -08:00 |
|
|
HAI
|
2ffe0a7363
|
Add get_amdgpu_memory_capacity() (#2049)
|
2024-11-15 22:51:48 -08:00 |
|
|
HAI
|
e5c6715003
|
Fix core (MI300X) with --enable-overlap (#2048)
|
2024-11-15 21:24:42 -08:00 |
|
|
HAI
|
b275ce0043
|
Github runner instructions for AMD (#2031)
|
2024-11-13 23:57:18 -08:00 |
|
|
HAI
|
087ab83223
|
[Performance, Triton] Optimize over mask compute to tl.load in fused_moe_kernel (#1980)
|
2024-11-10 18:54:43 -08:00 |
|
|
HAI
|
f9a377f650
|
[Release, ROCm] release ROCm docker build for AMD MI GPUs (#1957)
|
2024-11-08 00:14:15 -08:00 |
|
|
HAI
|
d32fba2a4d
|
[ENV, ROCm] update environment settings (#1939)
|
2024-11-07 18:24:36 -08:00 |
|
|
HAI
|
67c424cce3
|
[Performance, Triton Kernel Args] extend_attention, optimize kern args to _fwd_kernel (#1941)
|
2024-11-07 18:24:02 -08:00 |
|
|
HAI
|
dca87ec348
|
[Docs] fix 404 - Contributor Guide (#1942)
|
2024-11-07 16:50:45 +08:00 |
|
|
HAI
|
3cd2809277
|
[Docs, ROCm] update install to cover ROCm with MI GPUs (#1915)
|
2024-11-04 17:40:57 +08:00 |
|
|
HAI
|
d8e9d61f86
|
[Build, ROCm] Dockerfile.rocm for Instinct GPUs, with package updates (#1861)
|
2024-10-31 16:38:16 -07:00 |
|
|
HAI
|
2d4ce1b792
|
[Performance, Triton Kernel Args] _decode_grouped_softmax_reducev_fwd… (#1845)
|
2024-10-30 17:33:36 -07:00 |
|
|
HAI
|
5f65e2b830
|
[Performance, Hardware] MoE weights padding to AMD MI300x GPUs (#1836)
|
2024-10-30 12:17:32 -07:00 |
|
|
HAI
|
54dd3ea122
|
[FP8 KV Cache, Mixtral] Avoid KeyError at loading pre-quantized FP8 m… (#1835)
|
2024-10-29 13:58:03 -07:00 |
|
|
HAI
|
5010e0d2ca
|
[3rdparty, document] Add 3rdparty/amd, with profiling and tuning instructions to be added (#1822)
|
2024-10-29 10:51:02 -07:00 |
|
|
HAI
|
e11ab79e68
|
[Performance, hardware] MoE tuning update to AMD MI300x GPUs (#1619)
|
2024-10-10 22:48:15 -07:00 |
|
|
HAI
|
4d086719e5
|
[Bug] Fix decode stats error on output_len 1 (#1585)
|
2024-10-06 08:09:09 +00:00 |
|
|
HAI
|
e0b5dbcec1
|
[FP8 KV Cache] Avoid KeyError at loading pre-quantized FP8 model with kv_scale (#1559)
|
2024-10-03 01:52:26 -07:00 |
|
|
HAI
|
aa2750beb3
|
[Bugfix] Enable SGLang on AMD GPUs via PyTorch for ROCm (#1419) (#1453)
|
2024-09-18 02:01:35 -07:00 |
|
|
HAI
|
3a6e04185b
|
[Feature, Hardware] Enable SGLang on AMD GPUs via PyTorch for ROCm (#1420)
|
2024-09-17 07:43:52 +00:00 |
|