Wei Zhao 379689d533 [Perf] Support FP8 KV cache for Flashinfer MLA Sparse (#35891)

2026-03-07 13:51:54 -08:00

..

attention

[Perf] Support FP8 KV cache for Flashinfer MLA Sparse (#35891)

2026-03-07 13:51:54 -08:00

core

[V0 Deprecation] Remove unused swap_space parameter (#36216)

2026-03-07 22:09:55 +08:00

cudagraph

…

determinism

…

distributed

…

e2e

[Core] NGram GPU Implementation compatible with Async Scheduler (#29184)

2026-03-07 13:51:37 -08:00

ec_connector

…

engine

[V0 Deprecation] Remove unused swap_space parameter (#36216)

2026-03-07 22:09:55 +08:00

entrypoints

…

executor

…

kv_connector

[CI] Enable Crosslayer KV layout tests for ROCm platforms (#35416)

2026-03-07 13:49:56 -08:00

kv_offload

…

logits_processors

…

metrics

…

sample

…

shutdown

[CI] Fix startup error test (#36230)

2026-03-06 11:50:28 +00:00

spec_decode

[Bugfix] Fix block_size for hybrid model MTP (#36036)

2026-03-05 06:10:54 +00:00

streaming_input

…

structured_output

…

tracing

…

worker

[V0 Deprecation] Remove unused swap_space parameter (#36216)

2026-03-07 22:09:55 +08:00

__init__.py

…

test_oracle.py

…

test_outputs.py

…

test_request.py

…

test_serial_utils.py

…

utils.py

…