sunxin 33e01f22a8 [Feature][Sampling] Extend top-k_top-p sampling to all backends and unify greedy decoding with top_k=1 (#6894) ...

* update sampling

* fix

* fix

* fix mtp

* fix test

2026-03-19 01:43:10 -07:00

..

test_dummy_loader.py

[Loader] support dummy load weight (#6169)

2026-01-26 13:58:53 +08:00

test_load_attention.py

【FIX】Change the name of sparse attn from moba to plas (#4006) (#4076)

2025-09-23 10:26:40 +08:00

test_load_ernie_vl.py

[CI] Optimize port cleanup logic (#4860)

2025-11-06 19:13:48 +08:00

test_load_mtp.py

[Optimization] Support FA2/FA3/FA4 with attn_mask_q (#6354)

2026-02-05 14:39:00 +08:00

test_model_cache.py

[Optimization] Support FA2/FA3/FA4 with attn_mask_q (#6354)

2026-02-05 14:39:00 +08:00

test_offline_model.py

[Optimization] Support FA2/FA3/FA4 with attn_mask_q (#6354)

2026-02-05 14:39:00 +08:00

test_torch_model.py

[Feature][Sampling] Extend top-k_top-p sampling to all backends and unify greedy decoding with top_k=1 (#6894)

2026-03-19 01:43:10 -07:00

test_w4a8_model.py

Update test_w4a8_model.py (#4125)

2025-09-16 20:43:10 +08:00

utils.py

[CI] fix test_model_cache (#4982)

2025-11-12 20:26:49 +08:00