Lakonik

Distilled 4-step Qwen-Image models proposed in the paper: pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation Hansheng Chen 1 , Kai Zhang 2 , Hao Tan 2 , Leonidas Guibas 1 , Gordon Wetzstein 1 , Sai Bi 2 1 Stanford University, 2 Adobe Research We provide diffusers pipelines for easy inference. The following code demonstrates how to sample images from the distilled FLUX models. 4-NFE GM-Qwen (GMFlow Policy, Recommended) Note: GM-Qwen supports elastic inference. Feel free to set `numinferencesteps` to any value above 4.

Distilled 4-step and 8-step FLUX.1 models proposed in the paper: pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation Hansheng Chen 1 , Kai Zhang 2 , Hao Tan 2 , Leonidas Guibas 1 , Gordon Wetzstein 1 , Sai Bi 2 1 Stanford University, 2 Adobe Research We provide diffusers pipelines for easy inference. The following code demonstrates how to sample images from the distilled FLUX models. 4-NFE GM-FLUX (GMFlow Policy) Note: For the 8-NFE version, replace `gmfluxk8piid4step` with `gmfluxk8piid8step` and set `numinferencesteps=8`.