3 people 1417022c3d [ROCm][Kimi-K3] Optimize low-concurrency speculative KDA (#58045) ...

Signed-off-by: jiacao-amd <jiahui.cao@amd.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>

2026-09-25 12:33:30 -05:00

..

attention

[ROCm][Kimi-K3] Optimize low-concurrency speculative KDA (#58045)

2026-09-25 12:33:30 -05:00

core

[Bugfix][KV Cache][MLA] Align packed block strides for V3.2 sparse MLA (#55528)

2026-09-25 05:33:36 +00:00

engine

[Core] Model console logging as CLI configuration (#57205)

2026-09-25 11:36:59 +01:00

executor

[Core] Model console logging as CLI configuration (#57205)

2026-09-25 11:36:59 +01:00

fault_tolerance

[Feature] Add fault tolerance framework (simplified) for DP+EP external LB deployments (#44428)

2026-07-25 11:49:09 -04:00

hisparse

[Bugfix][HiSparse][NIXL] Import full blocks without tail prefill on D (#57049)

2026-09-17 15:26:10 -04:00

kv_hints

[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)

2026-09-21 09:56:11 -07:00

kv_offload

[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)

2026-09-21 09:56:11 -07:00

metrics

[Bugfix][Metrics] Do not log a 0.0% prefix cache hit rate before any query (#54990)

2026-09-17 18:19:13 +00:00

pool

[Pooling] Improve BGE-M3 sync pooling throughput by up to 3.13% (#53464)

2026-08-24 08:33:32 -04:00

sample

[Bugfix][LogitsProcessor] Validate ':' separator in custom logits processor FQCN (#56020)

2026-09-25 01:49:20 +00:00

simple_kv_offload

[Bugfix][KVConnector] Finalize saves on steps without a forward (#57775)

2026-09-25 00:06:36 +00:00

spec_decode

[Perf] Remove CPU-GPU sync in heterogeneous vocabulary speculative decoding (#57396)

2026-09-22 18:34:19 -04:00

structured_output

[Structured Outputs] Parse Lark grammars natively in the xgrammar backend (#58321)

2026-09-25 04:35:10 +00:00

watermarking

[MRV2][XPU] use xpu sample kernel in mrv2 sampler (#57277)

2026-09-21 13:24:41 +08:00

worker

[Bugfix] Stop allocator fragmentation from shrinking the KV cache during memory profiling (#58430)

2026-09-25 12:16:27 -04:00

__init__.py

…

cudagraph_dispatcher.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

kv_cache_interface.py

[Bugfix][KV Cache][MLA] Align packed block strides for V3.2 sparse MLA (#55528)

2026-09-25 05:33:36 +00:00

kv_cache_layout.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

kv_cache_spec_registry.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

outputs.py

[AuxOutput] Add block-keyed storage for routed-expert outputs (#45635)

2026-09-21 10:25:05 +00:00

request.py

[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)

2026-09-21 09:56:11 -07:00

serial_utils.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

utils.py

[Core][Logging] Fix JSON logging process decoration (#57957)

2026-09-25 10:12:28 -06:00