Michael Goin 7a06e5b05b [Bugfix] Fix MTP accuracy for GLM-5 (#34385) ...

Signed-off-by: mgoin <mgoin64@gmail.com>
(cherry picked from commit ec12d39d44)

2026-02-11 20:54:27 -08:00

..

attention

[XPU][9/N] clean up existing ipex code/doc (#34111)

2026-02-11 02:33:27 -08:00

core

[Bugfix][Core] Fix CPU memory leak from Request reference cycle in prefix caching (#34183)

2026-02-11 02:33:04 -08:00

engine

[Tiny] Rename encoder budget file to more specific name (#34103)

2026-02-09 03:48:19 +00:00

executor

[Feature] OTEL tracing during loading (#31162)

2026-02-05 16:59:28 -08:00

kv_offload

[Metrics] [KVConnector] Add Offloading Connector metrics (#27942)

2026-01-27 13:34:49 +00:00

metrics

[Metrics] Add labeled prompt token metrics for P/D disaggregation (#33290)

2026-02-04 07:46:48 +00:00

pool

[Refactor] Clean up pooler modules (#31897)

2026-01-08 00:07:43 +08:00

sample

[perf] v1/spec_decode: skip softmax for all-greedy rejection sampling (#32852)

2026-01-31 09:51:26 +00:00

spec_decode

[Bugfix] Fix MTP accuracy for GLM-5 (#34385)

2026-02-11 20:54:27 -08:00

structured_output

[Perf] Simplify DeepseekV32 tokenizer, ensure fast detokenization used (#33855)

2026-02-08 07:16:34 +00:00

worker

[Tiny] Rename encoder budget file to more specific name (#34103)

2026-02-09 03:48:19 +00:00

__init__.py

[V1] AsyncLLM Implementation (#9826)

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

cudagraph_dispatcher.py

Reduce the kernel overhead when num of active loras is smaller than max loras. Multiple cuda graphs are captured for each num of active-loras. (#32005)

2026-02-02 12:30:06 -05:00

kv_cache_interface.py

Make mypy opt-out instead of opt-in (#33205)

2026-01-29 09:12:26 +00:00

outputs.py

[ModelRunner V2] Fix spec decoding + logprobs (#33391)

2026-01-31 03:33:26 +00:00

request.py

[Bugfix][Core] Fix CPU memory leak from Request reference cycle in prefix caching (#34183)

2026-02-11 02:33:04 -08:00

serial_utils.py

[Multimodal] Simplify MM input definitions (#33331)

2026-01-29 13:32:04 +00:00

utils.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00