[Quant] 2-GPU ModelOpt FP8 DiT serve configs for Flux / Flux2-Klein / Qwen-Image / Z-Image#4207
Draft
david6666666 wants to merge 1 commit into
Draft
Conversation
…n, Qwen-Image, Z-Image DiT-only FP8 stage configs (TP=2) for the image-gen models whose ModelOpt FP8 adapter landed in vllm-project#2913. Split out of the video-gen calibration PR (vllm-project#3305) since these are image-gen scope. Class names verified against main's registry. Signed-off-by: lishunyang12 <lishunyang12@163.com>
15 tasks
Collaborator
hsliuustc0106
left a comment
hsliuustc0106
left a comment
There was a problem hiding this comment.
Config-only PR adding ModelOpt FP8 serve configs for 4 image-gen models (Flux, Flux2-Klein, Qwen-Image, Z-Image) with TP=2.
Please add verification evidence that these configs work (e.g., a test run or CI log) when ready.
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
DiT-only ModelOpt FP8 serve configs (TP=2, 2-GPU) for the image-gen models whose FP8 checkpoint adapter landed in #2913:
flux_dit_2gpu_fp8.yaml(FluxPipeline)flux2_klein_dit_2gpu_fp8.yaml(Flux2KleinPipeline)qwen_image_dit_2gpu_fp8.yaml(QwenImagePipeline)z_image_dit_2gpu_fp8.yaml(ZImagePipeline)Split out of the video-gen calibration PR (#3305) since these are image-gen scope. Each
model_class_nameis verified against main's diffusion registry. Configs only — no code changes.