RuohengMa de0c5e68fb [XPU] Split the block_attn operator into smaller operators (#6798) ...

* spliced block_attn

* adapt to latest vllm

* fix unit tests

* delete mtp+cudagraph 4 cards test

* fix vl model

* fix mtp

* fix slot mapping

2026-04-16 14:28:40 +08:00

..

run_ep4tp1_online.py

[XPU][CI]Rename test_ep4tp1_online.py to run_ep4tp1_online.py (#6805)

2026-03-12 16:16:20 +08:00

test_ep4tp4_all2all_cudagraph.py

Add PD+EP cudagraph Support

2026-03-12 13:20:59 +08:00

test_ep4tp4_all2all.py

[XPU] change XPU EP interface from xDeepEP to paddle (#5706)

2026-01-21 18:23:45 +08:00

test_ep4tp4_online.py

[XPU] change XPU EP interface from xDeepEP to paddle (#5706)

2026-01-21 18:23:45 +08:00

test_logprobs-topk_21b_tp4.py

[XPU] [CI] add xpu logprobs case (#6187)

2026-01-23 19:40:55 +08:00

test_logprobs-zmq_21b_tp4.py

[XPU] [CI] add xpu logprobs case (#6187)

2026-01-23 19:40:55 +08:00

test_mtp.py

[XPU][CI] XPU CI refactor (#6053)

2026-01-16 20:57:58 +08:00

test_pd_03b_tp1.py

[XPU][CI] Xpu ci update (#6089)

2026-01-19 16:09:09 +08:00

test_pd_21b_tp2_cudagraph.py

Add PD Cudagraph CI Case

2026-02-26 17:01:20 +08:00

test_pd_21b_tp2.py

[XPU][CI] Xpu ci update (#6089)

2026-01-19 16:09:09 +08:00

test_v1_mode.py

[XPU][CI] XPU CI refactor (#6053)

2026-01-16 20:57:58 +08:00

test_vl_model.py

[XPU][CI] XPU CI refactor (#6053)

2026-01-16 20:57:58 +08:00

test_w4a8_cudagraph.py

[XPU][CI]Add Cuda Graph CI Case (#6229)

2026-01-26 23:20:44 +08:00

test_w4a8.py

[XPU][CI] XPU CI refactor (#6053)

2026-01-16 20:57:58 +08:00

test_w16a16_21b_tp4.py

[XPU] support warmup with ep & remove apply_tp_fused_op (#6289)

2026-02-28 15:40:36 +08:00