Add PAI-Bench-C reproduction guide for Cosmos3#224
Open
trungtpham wants to merge 1 commit into
Open
Conversation
trungtpham
force-pushed
the
features/paibenchc-reproduce
branch
14 times, most recently
from
80a3c53 to
4981e86
Compare
qianlim
reviewed
Jun 22, 2026
trungtpham
force-pushed
the
features/paibenchc-reproduce
branch
4 times, most recently
from
b0233cc to
f86e7ff
Compare
Contributor
Author
|
Paibench-c results on demo 4 samples run by @trungtpham |
Contributor
Author
|
@lfengad please help review this PR? Thanks |
trungtpham
force-pushed
the
features/paibenchc-reproduce
branch
from
f86e7ff to
cf2e416
Compare
|
QA evaluation report can be found at https://docs.google.com/document/d/1B6YkXoz9jnh4CBZqCuptLegHhqkorpa8BTQfKVkoP7w/edit?tab=t.ridcxjs8xsjg |
- Self-contained run_paibench_c.sh for end-to-end generation and evaluation - Replaces PAIBENCH_C_DOWNLOAD_MODE/DEMO_TASK/DEMO_NUM_TASKS with PAIBENCH_C_NUM_SAMPLES - Defaults to 4-task smoke-test; set to 600 for the full benchmark - Includes SKIP_GEN flag, multi-GPU support, and per-checkpoint output dirs - Fixes: OOM for Super model, output overwrite between Nano/Super runs - Fixes: NCCL barrier timeout, CUDA OOM (SAM detection cap), broken venv detection - Aligns evaluation with internal Cosmos pipeline (GT depth/seg on-the-fly, raw video bytes for SAM, Canny RGB->GRAY conversion fix) - Uses trungtpham/pai-bench-c-cosmos3 fork with upstream bug fixes
trungtpham
force-pushed
the
features/paibenchc-reproduce
branch
from
cf2e416 to
5e923a2
Compare
qmiao-hub
approved these changes
Jun 27, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds an end-to-end recipe for reproducing PAI-Bench Conditional Generation (PAI-Bench-C) results with Cosmos3 using the native Cosmos Framework PyTorch entrypoint.
New files under evaluation/cosmos3/generator/paibench_c/:
PAIBench-C Reproducibility
