Michael GoinandCodex a7d00ec051 [Bugfix] DFlash fc sized wrong when num_target_layers != num_hidden_layers (#48524) ...

Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Codex <noreply@openai.com>

2026-07-21 14:42:52 -07:00

..

autoregressive

[Bugfix][Spec-Decode] Populate draft seq_lens_cpu_upper_bound for spec-decode attention metadata (#44492)

2026-07-20 20:50:58 +00:00

dflash

Support loading sample_from_anchor flag from speculators config (#48639)

2026-07-20 14:52:43 -07:00

dspark

Support loading sample_from_anchor flag from speculators config (#48639)

2026-07-20 14:52:43 -07:00

eagle

[Bugfix] DFlash fc sized wrong when num_target_layers != num_hidden_layers (#48524)

2026-07-21 14:42:52 -07:00

gemma4

[Model Runner V2][Spec Decode] Handle tuple hidden states from MTP draft models (#46786)

2026-06-27 18:38:07 -07:00

mtp

[Model Runner V2][Spec Decode] Handle tuple hidden states from MTP draft models (#46786)

2026-06-27 18:38:07 -07:00

__init__.py

[Spec Decode] DSpark (#46995)

2026-07-01 08:30:24 -07:00

rejection_sampler_utils.py

[Bugfix][Model Runner V2][Spec Decode] Fix int32 offset overflow in block verification kernels (#47383)

2026-07-02 07:32:38 -07:00

rejection_sampler.py

[Bugfix][V1/V2] Fix prompt_logprobs to respect logprobs_mode (#47680)

2026-07-17 21:58:59 +01:00

speculator.py

[Bugfix][Spec-Decode] Populate draft seq_lens_cpu_upper_bound for spec-decode attention metadata (#44492)

2026-07-20 20:50:58 +00:00

utils.py

[Perf] Use blocking CUDA events to avoid busy polling cuda driver lock (#47081)

2026-07-07 09:36:10 +00:00