Repository navigation
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
MOSS-Image's official runtime requires newer Torch and Transformers versions that conflict with SSUI's shared environment. This ports its text-only Qwen3.5 encoder, DiT, FLUX.2 VAE and sampling schedule into
backend/moss_image/, allowing Base/SFT bundles to load without a separate environment or global dependency changes.runner.tempworkflow expression. The optional GPU workflow remains opt-in/scheduled; no full model download is added to normal CI.Validation: the synthetic-weight end-to-end test exercises HTTP requests, real loopback WebSockets, sandbox execution, sharded component loading and PNG output. It checks prepare without weights, deterministic seeds, layout input, output uniqueness and failure recovery. Numerical reference, node contract, discovery, workflow compiler, Diffusers compatibility and FLUX.2 regressions were also checked with Torch 2.4.1+cpu / Transformers 4.55.4 / Diffusers 0.38.0. Local
pip checkpasses.Full Base/SFT weights and GPU hardware were intentionally not tested. Component offload is implemented; the official minimal/hybrid low-memory mode is not included. Existing dependency constraints and platform lockfiles are unchanged.