Files

T

Baorun (Lauren) Mu f85e479e66 [Feature] ViT Full CUDA Graph (#35963 )

Signed-off-by: Baorun Mu <bmu@nvidia.com>

2026-03-23 13:01:10 +08:00

metrics

[Model Runner V2] Support num NaNs in logits (#30187 )

2025-12-09 10:00:49 -08:00

[Feature] ViT Full CUDA Graph (#35963 )

2026-03-23 13:01:10 +08:00

model_states

[Model Runner V2] Support Streaming Inputs (#37028 )

2026-03-20 20:42:25 +00:00

pool

[Model Runner V2] Do not initialize sampler for non-last PP ranks (#36824 )

2026-03-12 03:55:28 +00:00

sample

[MRV2] Use FP64 for Gumbel noise (#37798 )

2026-03-22 12:28:10 -07:00

spec_decode

[MRV2] Use FP64 for Gumbel noise (#37798 )

2026-03-22 12:28:10 -07:00

__init__.py

GPU Model Runner V2 (#25266 )

2025-11-21 08:20:55 -08:00

async_utils.py

[Model Runner V2] Fix pooling (#36019 )

2026-03-04 10:53:17 -08:00

attn_utils.py

[Model Runner V2] Support Streaming Inputs (#37028 )

2026-03-20 20:42:25 +00:00

block_table.py

[MRV2] Avoid recompilation of _gather_block_tables_kernel (#37645 )

2026-03-20 10:31:45 -07:00

buffer_utils.py

[V1] Remove pin_memory() in async_copy_to_gpu to fix sporadic stalls (#37006 )

2026-03-14 01:37:32 +00:00

cp_utils.py

[Model Runner V2] Minor simplification for DCP (#34786 )

2026-02-18 11:04:53 -08:00

cudagraph_utils.py

[Model Runner V2] Enable piecewise CUDA graphs for pipeline parallelism (#35162 )

2026-03-22 20:48:25 +00:00

dp_utils.py

[MRV2] Extensible CG dispatch rework (#35959 )

2026-03-09 13:58:45 -07:00

input_batch.py

[Model Runner V2] Do not initialize sampler for non-last PP ranks (#36824 )

2026-03-12 03:55:28 +00:00

kv_connector.py

[kv_offload+HMA][0/N]: Support block-level preemption handling (#34805 )

2026-03-18 08:49:53 +02:00

lora_utils.py

[ModelRunner V2] Misc minor simplifications and optimizations (#33467 )

2026-02-01 22:17:14 +00:00

model_runner.py

[Model Runner V2] Enable piecewise CUDA graphs for pipeline parallelism (#35162 )

2026-03-22 20:48:25 +00:00

pp_utils.py

[Model Runner V2] A bit more PP simplification (#34766 )

2026-02-17 21:39:07 -08:00

README.md

GPU Model Runner V2 (#25266 )

2025-11-21 08:20:55 -08:00

states.py

[Model Runner V2] Support Streaming Inputs (#37028 )

2026-03-20 20:42:25 +00:00

structured_outputs.py

[ModelRunner V2] Misc minor simplifications and optimizations (#33467 )

2026-02-01 22:17:14 +00:00

warmup.py

[Model Runner V2] Fix warmup for pipeline parallel (#36280 )

2026-03-06 16:58:51 -08:00

README.md

[Experimental] Model Runner V2

This directory contains the new model runner which is under active development. Ping Woosuk Kwon for any changes.