Enable self refiner modality guidance#120
Open
Fabio Maia (fabiormoura) wants to merge 4 commits intoLightricks:mainfrom
Open
Enable self refiner modality guidance#120Fabio Maia (fabiormoura) wants to merge 4 commits intoLightricks:mainfrom
Fabio Maia (fabiormoura) wants to merge 4 commits intoLightricks:mainfrom
Conversation
The LTX-2 fork's X0Model.forward() accepts single Modality objects, not batched lists like Wan2GP's modified transformer. Use two separate forward passes (normal + cross-attn perturbed) instead of batching. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The Triton fused_add_round_kernel uses fp8e4nv, which requires Ada Lovelace (compute capability >= 8.9). On Ampere (sm_86, e.g. A40), this fails with 'type fp8e4nv not supported in this architecture'. Detect GPU compute capability at first call and fall back to pure PyTorch: add in bfloat16 then deterministic cast to FP8. Slightly less accurate than stochastic rounding but equivalent for inference.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
No description provided.