Chengji Yao a454748544

[TPU][V1] Refine tpu_model_runner to mitigate future recompilation issues (#16275) ...

Signed-off-by: Chengji Yao <chengjiyao@google.com>

2025-04-09 18:51:51 -06:00

..

core

[Feature] Estimate max-model-len use available KV cache memory (#16168)

2025-04-08 19:12:51 -07:00

e2e

[V1] Implement sliding window attention in kv_cache_manager (#14097)

2025-04-01 00:33:17 -07:00

engine

[V1][BugFix] Exit properly if engine core fails during startup (#16137)

2025-04-07 15:30:15 -07:00

entrypoints

[V1] Fix json_object support with xgrammar (#15488)

2025-04-02 02:00:08 -07:00

sample

[V1][Sampler] Faster top-k only implementation (#15478)

2025-03-26 10:56:47 -07:00

spec_decode

[V1][Spec Decode] Respect prompt_lookup_max (#15348)

2025-03-23 10:41:44 -07:00

structured_output

[CI] xgrammar structured output supports Enum. (#15757)

2025-03-29 20:20:02 -07:00

tpu

[TPU][V1] Refine tpu_model_runner to mitigate future recompilation issues (#16275)

2025-04-09 18:51:51 -06:00

worker

[V1] Scheduler Refactoring [1/N] - Add Scheduler Interface (#15250)

2025-03-20 17:50:43 -07:00

__init__.py

[V1] AsyncLLM Implementation (#9826)

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

test_async_llm_dp.py

[V1] AsyncLLM data parallel (#13923)

2025-03-27 16:14:41 -07:00

test_oracle.py

[Misc] Enable V1 LoRA by default (#15320)

2025-04-01 16:53:56 +08:00

test_stats.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

test_utils.py

Update deprecated Python 3.8 typing (#13971)

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