Files

T

Woosuk Kwonandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com> 0a3e2dbc09 [Optimization] Skip DP padding tokens in MoE (#46428 )

Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

2026-06-23 14:54:46 -07:00

metrics

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

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

[Bugfix][Spec Decode] Fix EAGLE drafter multimodal encoder cache misses (#46315 )

2026-06-22 18:14:02 +00:00

model_states

[MRV2] Generalize use of WhisperModelState (#46096 )

2026-06-22 12:40:02 -07:00

pool

[Frontend] Re-enable running MaxSim on GPU (#38620 )

2026-04-03 00:03:13 +08:00

sample

[Bugfix][Model Runner V2] Preserve all allowed_token_ids in the logit bias kernel (#46245 )

2026-06-23 07:01:13 +00:00

spec_decode

[Model Runer V2][DFlash] Fix lm head sharing for dflash (#46435 )

2026-06-23 19:09:06 +00:00

__init__.py

GPU Model Runner V2 (#25266 )

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

async_utils.py

[BugFix][MRV2] Fix cuda event reuse race (#39115 )

2026-04-07 00:21:09 +00:00

attn_utils.py

[Spec Decode] Support mixed KV page sizes for DFlash (#45181 )

2026-06-21 22:45:14 +08:00

block_table.py

[PERF] Fuse multi-group block table staged writes (#44944 )

2026-06-16 10:53:27 -07:00

buffer_utils.py

[Core] Ensure memory is pinned prior to async h2d copy (#45424 )

2026-06-20 20:02:24 -07:00

cp_utils.py

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

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

cudagraph_utils.py

[Optimization] Skip DP padding tokens in MoE (#46428 )

2026-06-23 14:54:46 -07:00

dp_utils.py

[Model Runner V2][Bugfix] Fix MRV2 LoRA warmup (#35536 )

2026-06-15 13:17:23 -07:00

eplb_utils.py

[Bugfix] Unwrap VLM wrappers for EPLB on Model Runner V2 (#42706 )

2026-05-16 07:20:33 +08:00

input_batch.py

[Optimization] Skip DP padding tokens in MoE (#46428 )

2026-06-23 14:54:46 -07:00

kv_connector.py

[Core] Cleanup KVConnector handling with PP + fix MRV2 (#43732 )

2026-05-28 13:12:03 -07:00

lora_utils.py

[Model Runner V2][Bugfix] Fix MRV2 LoRA warmup (#35536 )

2026-06-15 13:17:23 -07:00

model_runner.py

[Optimization] Skip DP padding tokens in MoE (#46428 )

2026-06-23 14:54:46 -07:00

pp_utils.py

[ModelRunnerV2] Avoid pipeline parallel bubbles (#42187 )

2026-06-02 14:02:01 -07:00

README.md

GPU Model Runner V2 (#25266 )

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

shutdown.py

[MRV2] Add shutdown() method (#41297 )

2026-05-02 21:06:30 -07:00

states.py

[ModelRunnerV2] Avoid pipeline parallel bubbles (#42187 )

2026-06-02 14:02:01 -07:00

structured_outputs.py

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

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

warmup.py

Enable DeepSeek V4 and GLM-5.1 on SM120 (#43477 )

2026-06-22 11:54:14 -07:00

README.md

[Experimental] Model Runner V2

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