Woosuk Kwon 37478c18cf async output ...

Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>

2025-09-19 07:10:42 +00:00

..

attention

fix

2025-09-18 14:52:23 -07:00

core

Merge branch 'main' into woosuk/model-runner-v2

2025-09-19 06:52:46 +00:00

engine

[CORE] Prompt Embeddings Support for v1 Engine (#24278)

2025-09-19 08:03:09 +08:00

executor

[P/D]kv_output_aggregator support P TP > D TP (#23917)

2025-09-15 11:36:06 +02:00

kv_offload

[KV offload][2/N] Introduce LRU-based CPU offloading management (#20075)

2025-09-19 00:20:51 +00:00

metrics

[Metrics] Hide deprecated metrics with gpu_ prefix (#24245)

2025-09-15 20:15:57 -06:00

pool

[Performance] V1 Pooling Models E2E Performance Optimization (#23162)

2025-08-21 13:26:09 +00:00

sample

[CORE] Prompt Embeddings Support for v1 Engine (#24278)

2025-09-19 08:03:09 +08:00

spec_decode

[Spec Decode] Efficient padded speculation (#24539)

2025-09-18 01:07:24 -04:00

structured_output

[Structured Output][Refactor] Move apply_grammar_bitmask() method from ModelRunner to structured output utils (#21999)

2025-09-18 20:44:31 +08:00

worker

async output

2025-09-19 07:10:42 +00:00

__init__.py

[V1] AsyncLLM Implementation (#9826)

2024-11-11 23:05:38 +00:00

cudagraph_dispatcher.py

[CI] Fix linting error on main (#23835)

2025-08-28 06:52:01 -07:00

kv_cache_interface.py

merge

2025-09-18 12:13:56 -07:00

outputs.py

logprobs

2025-09-18 17:23:02 -07:00

request.py

[CORE] Prompt Embeddings Support for v1 Engine (#24278)

2025-09-19 08:03:09 +08:00

serial_utils.py

[CORE] Prompt Embeddings Support for v1 Engine (#24278)

2025-09-19 08:03:09 +08:00

utils.py

Allow users to specify kv cache memory size (#21489)

2025-09-11 13:41:07 +00:00