3 people 1356df53bd [Kernel] Use flash-attn for decoding (#3648) ...

Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: LiuXiaoxuanPKU <lilyliupku@gmail.com>

2024-05-13 15:50:33 -07:00

..

__init__.py

Change the name to vLLM (#150)

2023-06-17 03:07:40 -07:00

cache_engine.py

[Misc] Enhance attention selector (#4751)

2024-05-13 10:47:25 -07:00

cpu_model_runner.py

[Misc] Enhance attention selector (#4751)

2024-05-13 10:47:25 -07:00

cpu_worker.py

[Misc] Enhance attention selector (#4751)

2024-05-13 10:47:25 -07:00

embedding_model_runner.py

[Misc] Enhance attention selector (#4751)

2024-05-13 10:47:25 -07:00

model_runner.py

[Kernel] Use flash-attn for decoding (#3648)

2024-05-13 15:50:33 -07:00

neuron_model_runner.py

[Core][Model runner refactoring 1/N] Refactor attn metadata term (#4518)

2024-05-03 10:20:12 -07:00

neuron_worker.py

[Core] RayWorkerVllm --> WorkerWrapper to reduce duplication (#4024)

2024-04-17 08:34:33 +00:00

worker_base.py

[Misc][Refactor] Introduce ExecuteModelData (#4540)

2024-05-03 17:47:07 -07:00

worker.py

[Core][Distributed] refactor custom allreduce to support multiple tp groups (#4754)

2024-05-12 17:47:59 -07:00