Antoni Baum 9841d48a10 Use TGI-like incremental detokenization (#984)

2023-09-13 13:38:01 -07:00

..

async_engine

Start background task in AsyncLLMEngine.generate (#988)

2023-09-08 00:03:39 -07:00

engine

Use TGI-like incremental detokenization (#984)

2023-09-13 13:38:01 -07:00

kernels

Use FP32 in RoPE initialization (#1004)

2023-09-11 00:26:35 -07:00

models

Add tests for models (#922)

2023-09-01 11:19:43 +09:00

samplers

Align vLLM's beam search implementation with HF generate (#857)

2023-09-04 17:29:42 -07:00

conftest.py

Use queue for finished requests (#957)

2023-09-05 19:27:23 -07:00