Nicolò Lucchesi b3f2fddd17

[TPU][V1] Fix exponential padding when max-num-batched-tokens is not a power of 2 (#16596) ...

Signed-off-by: NickLucche <nlucches@redhat.com>

2025-04-14 17:01:05 +00:00

..

worker

[TPU][V1] Fix exponential padding when max-num-batched-tokens is not a power of 2 (#16596)

2025-04-14 17:01:05 +00:00

__init__.py

[V1] TPU - Add tensor parallel support via Ray (#13618)

2025-03-08 08:19:38 -05:00

test_basic.py

[V1] TPU CI - Add basic perf regression test (#15414)

2025-03-31 13:25:20 -04:00

test_mha_attn.py

[TPU][V1] MHA Pallas backend (#15288)

2025-03-21 08:50:39 -07:00

test_pallas.py

[TPU] Update PyTorch/XLA (#16288)

2025-04-09 14:46:32 +08:00

test_perf.py

[V1] TPU CI - Add basic perf regression test (#15414)

2025-03-31 13:25:20 -04:00

test_sampler.py

[TPU][V1] Disable per-request seed/Generator (#16172)

2025-04-10 17:05:44 -04:00

test_topk_topp_sampler.py

[V1][TPU] TPU-optimized top-p implementation (avoids scattering). (#15736)

2025-04-02 17:18:08 -07:00