tzhouamandClaude Fable 5.1 74acd66307 [Edge][WP7] Add CudaIpcConnector (CUDA IPC for device tensor leaves over shm control plane) ...

Wraps SharedMemoryConnector: CUDA tensor leaves are exported with
torch.multiprocessing reduce_tensor + an inter-process CUDA event, rebuilt and
cloned by the consumer, released by the producer after an ack file appears.
CPU leaves and metadata keep the shm path. Registered as "CudaIpcConnector";
deploy/qwen3_tts_cuda_ipc.yaml selects it for the talker->code2wav hop.
Adoption in the edge profile is measurement-gated (hop latency bench).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-09 14:41:33 +08:00

..

assets

[Feat] Add support and testcase for HunyuanImage-3.0-Instruct-Distil (#4048)

2026-09-03 02:43:13 +00:00

attention

[Rebase] Rebase to vllm 0.26.0 (#5443)

2026-07-27 23:35:49 +08:00

benchmarks

[Edge][WP4] Add hardware adaptation: probe, derived overrides, auto profile, calibration

2026-09-09 11:01:40 +08:00

buildkite

[CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543)

2026-09-02 15:39:49 +00:00

clients

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

config

[Edge][WP6] Slim the AR/TTS import path and add an edge dependency set

2026-09-09 11:14:50 +08:00

core

[Model] Preserve MiniCPM-o Talker full attention until capacity (#6799)

2026-09-06 22:25:35 +08:00

dfx

[CI][Bugfix] Use the matching checkpoint for Qwen3-TTS VoiceDesign tests (#7054)

2026-09-06 04:27:21 +00:00

diffusion

[Bugfix][NPU] Fix MiniMax-H3 INT8 quantization dispatch (#6876)

2026-09-07 14:50:44 +00:00

distributed

[Edge][WP7] Add CudaIpcConnector (CUDA IPC for device tensor leaves over shm control plane)

2026-09-09 14:41:33 +08:00

docs

[Bugfix] Stop reading the removed OmniRequestOutput.request_output accessor (#6172)

2026-08-15 07:07:28 +08:00

e2e

[Edge][WP2] Finalize edge profile init knobs from init_profile; WP4 CPU emulation results

2026-09-09 14:40:13 +08:00

edge

[Edge][WP4] Add hardware adaptation: probe, derived overrides, auto profile, calibration

2026-09-09 11:01:40 +08:00

engine

[Edge][WP2] Add initialization timeline and cold-start profiler

2026-09-09 11:01:39 +08:00

entrypoints

[Edge][WP5] Add codec-token streaming mode with on-device decoding

2026-09-09 11:10:13 +08:00

examples

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

helpers

[Edge][WP1] Add edge deploy profile and deploy_profile selector

2026-09-09 10:27:33 +08:00

host_weight_runtime

[Bugfix] Make HWR filesystem lifecycle transitions fail-closed (#6692)

2026-08-28 09:30:17 +08:00

inputs

[Model][Feat] Support Minimax-H3 quality grading requests through dynamic loading/unloading with Cache-DiT (#5853)

2026-08-07 09:07:23 +00:00

metrics

[Bugfix] Report engine-queued requests as waiting in vllm_omni:num_requests gauges (#6549)

2026-08-29 03:58:26 +00:00

model_executor

[Model] Preserve MiniCPM-o Talker full attention until capacity (#6799)

2026-09-06 22:25:35 +08:00

model_extras

[Model] Add native MAGI-2 Preview diffusion support (#5918)

2026-09-05 15:32:30 +08:00

model_tests

[Model] Add native MAGI-2 Preview diffusion support (#5918)

2026-09-05 15:32:30 +08:00

platforms

[Edge][WP3] Add CPU OmniPlatform (AR + generation stages on vLLM's CPU backend)

2026-09-09 11:02:14 +08:00

profile

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

quantization

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

reasoning

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

tools

[Migrate] Move sampling parameter overrides from legacy dispatches to TTS adapters (#5272)

2026-08-16 21:14:26 -07:00

utils

[Core] Validate PipelineConfig in __post_init__ (#6829)

2026-09-01 11:17:57 +08:00

worker

[Edge][WP3] Add CPU OmniPlatform (AR + generation stages on vLLM's CPU backend)

2026-09-09 11:02:14 +08:00

__init__.py

[Fix] Resolve comments & update vLLM-Omni name usages. (#122)

2025-12-01 08:05:24 +08:00

conftest.py

[CI]Split GPU jobs by cards_* and reject hand-written SKU marks (#6650)

2026-08-29 18:42:33 +08:00