Wallas Henrique c27df94e1f

[Bugfix] Fix chunked prefill with model dtype float32 on Turing Devices (#9850) ...

Signed-off-by: Wallas Santos <wallashss@ibm.com>
Co-authored-by: Michael Goin <michael@neuralmagic.com>

2024-11-25 12:23:32 -05:00

..

backends

[Bugfix] Avoid import AttentionMetadata explicitly in Mllama (#10593)

2024-11-23 18:12:20 +00:00

ops

[Bugfix] Fix chunked prefill with model dtype float32 on Turing Devices (#9850)

2024-11-25 12:23:32 -05:00

__init__.py

[Core] Add AttentionState abstraction (#7663)

2024-08-20 18:50:45 +00:00

layer.py

[misc] move functions to config.py (#10624)

2024-11-25 09:27:30 +00:00

selector.py

[Platform][Refactor] Extract func get_default_attn_backend to Platform (#10358)

2024-11-19 11:22:26 +08:00