Woosuk Kwon 41fb013d29 [V1][Spec Decode] Always use argmax for sampling draft tokens (#16899) ...

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

2025-04-23 14:57:43 -07:00

..

__init__.py

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

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

block_table.py

Update deprecated Python 3.8 typing (#13971)

2025-03-02 17:34:51 -08:00

gpu_input_batch.py

[V1][Spec Decode] Implement Eagle Proposer [1/N] (#15729)

2025-04-01 12:33:16 -07:00

gpu_model_runner.py

[V1][Spec Decode] Always use argmax for sampling draft tokens (#16899)

2025-04-23 14:57:43 -07:00

gpu_worker.py

Restore buffers when wake up from level 2 sleep (#16564) (#16889)

2025-04-21 20:18:28 +08:00

lora_model_runner_mixin.py

[Kernels] LoRA - Retire SGMV and BGMV Kernels (#14685)

2025-03-18 09:47:53 +00:00

tpu_model_runner.py

[BugFix][V1] Fix int32 token index overflow when preparing input ids (#16806)

2025-04-23 12:12:35 -07:00

tpu_worker.py

[TPU][V1] Capture multimodal encoder during model compilation (#15051)

2025-04-21 18:36:59 -06:00

utils.py

[V1] Scatter and gather placeholders in the model runner (#16076)

2025-04-08 10:43:41 +08:00

worker_base.py

[v1] Refactor KVCacheConfig (#14079)

2025-03-21 04:56:27 -07:00