Skip to content

docs(skills): add Cosmos transfer inference skill - #74

Open
chrisvoncsefalvay wants to merge 12 commits into
NVIDIA:mainfrom
chrisvoncsefalvay:feat/cosmos-transfer-inference-skill
Open

docs(skills): add Cosmos transfer inference skill#74
chrisvoncsefalvay wants to merge 12 commits into
NVIDIA:mainfrom
chrisvoncsefalvay:feat/cosmos-transfer-inference-skill

Conversation

@chrisvoncsefalvay

Copy link
Copy Markdown

Teaches agents how to run Cosmos3 transfer inference with one or more conditioning controls (blur, edge, depth and so on).

@lfengad

lfengad commented Jul 2, 2026

Copy link
Copy Markdown
Collaborator

@trungtpham could have a review and check? THX

@lfengad lfengad left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks a lot for putting this together! 🙏 This is genuinely useful — docs/inference.md has zero transfer coverage today, so this skill basically becomes the only documentation we have for the feature. The structure, frontmatter, and the .agents + .claude duplication all match our existing skills, so that part is spot on.

I went through it line by line against transfer.py / args.py / inference.py, and most of the "Source facts" check out exactly — hint enum ordering, weight normalization, single-hint-only defaults, the output filenames, the no-mixing / no-batching asserts, JSON prompt compaction, and single-frame metadata stripping are all accurate. Nice work on that.

I did bump into a handful of things while reading, left inline below. Two of them (the negative_prompt_file paths and the aspect_ratio rule) I think would actually break or silently mislead someone following the skill, so those are the ones I'd love to see addressed. The rest are smaller / take-them-or-leave-them.

One housekeeping note: I only commented on the .agents/ copy to avoid duplicating everything, but every comment applies to the identical .claude/skills/cosmos3-transfer-inference/SKILL.md too.

Happy to help with any of these if it's easier — just let me know!

Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Comment thread .agents/skills/cosmos3-transfer-inference/SKILL.md Outdated
Clarify transfer paths, examples, launch settings, output composition, and batching constraints. Correct stale multi-GPU latency guidance while retaining full-world FSDP for Super.

Signed-off-by: Chris von Csefalvay <chris@chrisvoncsefalvay.com>
@chrisvoncsefalvay

Copy link
Copy Markdown
Author

Thank you again for such a thoughtful review, @lfengad. I’ve addressed all 14 threads in f7d0c29. The two skill copies remain byte-identical, all fenced JSON examples parse, and the targeted pre-commit and link checks pass.

Beyond the narrowly required correction to the existing parallelism guidance, I have not added a broader transfer section to docs/inference.md because I did not want to overburden this PR. If you think that documentation would be better added here than in a separate PR, I am happy to document transfer properly in this PR.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants