Woosuk Kwon a79b122400 [V1] Do not allocate beyond the max_model_len (#10730) ...

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

2024-11-28 00:13:15 -08:00

..

__init__.py

[V1] Implement vLLM V1 [1/N] (#9289)

2024-10-22 01:24:07 -07:00

encoder_cache_manager.py

[V1] Support VLMs with fine-grained scheduling (#9871)

2024-11-13 04:53:13 +00:00

kv_cache_manager.py

[V1] Do not allocate beyond the max_model_len (#10730)

2024-11-28 00:13:15 -08:00

kv_cache_utils.py

[v1] Refactor KVCacheManager for more hash input than token ids (#10507)

2024-11-22 23:27:25 +00:00

scheduler.py

[V1] Do not allocate beyond the max_model_len (#10730)

2024-11-28 00:13:15 -08:00