Commit Graph
20 Commits
Author SHA1 Message Date
Huamin Li 157722da75 [perf] Use pinned memory for async H2D transfer in do_mamba_copy_block (#35480)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2026-02-28 01:50:37 +08:00
Huamin LiandCyrus Leung 1fe462168c [perf] Avoid dtype promotion sync in mamba_get_block_table_tensor (#34870)
Signed-off-by: Huamin Li <3ericli@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
2026-02-20 06:21:56 -08:00
Huamin Li 83805a6078 [CI] Skip paddleocr_vl for transformer 4.57.3 (#29758)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-12-01 04:38:06 +00:00
Huamin Li 82c795d6f2 Fix AttributeError about _use_fi_prefill (#29734)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-11-30 06:04:55 +00:00
Huamin Li 3fd1fb0b60 Revert "[LoRA] Support FusedMoE LoRA Triton kernel for mxfp4 (#28971)" (#29697)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-11-28 15:26:52 -08:00
Huamin Li 70d5953f82 Revert "[Bugfix] Fix GPT-OSS AR+NORM fusion (#28841)" (#29483)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-11-26 22:27:26 +08:00
8ac3a41487 [CI Failure] Fix Gemma3 RoPE configuration for sliding attention layers (#29111)
Signed-off-by: Huamin Li <3ericli@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
2025-11-20 23:53:30 -08:00
Huamin Li 07a606aa7e [CI Failure] Fix backend selection for encoder-only models (#28534)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-11-13 10:11:27 -05:00
Huamin Li c748355e0d [CI] Introduce autorun_on_main feature (#27836)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-11-12 08:51:19 +00:00
Huamin Li 933cdea440 [BugFix] Don’t compute reorder threshold when there are no attention groups (#27861) 2025-10-31 11:36:18 +00:00
Huamin Li 1994de99ea [CI Failure] Fix test_kv_cache_model_load_and_run (#27717)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-30 12:27:53 +00:00
Huamin Li c7d2a554ba [CI Failure] fix test_default_mm_loras (#27795)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-30 18:13:03 +08:00
Huamin Li 5be1bed790 [CI/Build]Add eval config for Qwen3-235B-A22B-Instruct-2507-FP8 (#27113)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-30 07:50:56 +00:00
Huamin LiandYe Qi c312320764 [CI/Build] tests(v1): feed Triton attention the (num_blocks, 2, …) KV cache layout in backend-correctness tests (#26663)
Signed-off-by: Huamin Li <3ericli@gmail.com>
Co-authored-by: Ye (Charlotte) Qi <yeq@meta.com>
2025-10-17 21:11:26 -07:00
Huamin Li 87efc681db llama4_vision_rope: add HIP override to accept (q, k) and avoid (positions, q, k) mismatch (#26790)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-14 11:54:12 -07:00
Huamin Li 4fca1a1bd2 [easy] fix pre commit error on trunk (#26665)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-12 21:25:34 +00:00
Huamin Li 0cd103e7cb CP: make correct_attn_out robust to 4‑D views and fix Triton arg binding (#26509)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-11 20:50:57 +00:00
Huamin Li 7d6b03381e [CI Failure] fix_test_auto_prefix_cache_support (#26053)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-04 02:44:49 -07:00
Huamin Li c36f0aa300 Fix test_mamba_ssm_ssd.py due to missing _query_start_loc_to_chunk_indices_offsets (#25995)
Signed-off-by: Huamin Li <3ericli@gmail.com>
2025-10-01 18:18:36 +00:00
Huamin LiandLu Fang 62b38dc832 [Doc] improve test-pipeline.yaml documentation (#25305)
Signed-off-by: Huamin Li <3ericli@gmail.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
2025-09-20 20:29:12 -07:00