aredden/flux-fp8-api

Issues

• After changing lora many times, the pictures are getting weirder and weirder

  #31 opened 2 months ago by 81549361
  3

• Initial Delay in Image Generation with Flux Schnell on H100

  #24 opened 3 months ago by uayodev
  6

• The possibility of supporting GPUs with other architectures

  #33 opened 2 months ago by ziyaxuanyi
  1

• when load certain lora, AttributeError: 'Flux' object has no attribute 'diffusion_model' happened.

  #34 opened 2 months ago by fyepi
  1

• Certain lora not applied correctly.

  #36 opened 2 months ago by fyepi
  1

• Acceleration not as expected

  #35 opened 2 months ago by alecyan1993
  1

• Compatibility Inquiry: Using flux-fp8 with OpenFLUX.1

  #32 opened 2 months ago by veyorokon
  1

• Where is the code about "remaining layers use faster half precision accumulate"?

  #10 opened 4 months ago by goldhuang
  5

• Issue: torch._scaled_mm RuntimeError on RTX 6000 (with runpod/pytorch:2.4.0-py3.11-cuda12.4.1-devel-ubuntu22.04)

  #30 opened 2 months ago by veyorokon
  2

• LoRA loaded successfully but the effect wasn't applied

  #28 opened 2 months ago by EntroSanity
  2

• A question regarding whether the LoRA has been successfully applied to the inference process？

  #29 opened 2 months ago by zhangqi420
  0

• Any plans for controlnet + inpainting support?

  #15 opened 4 months ago by 0xtempest
  2

• Potential LoRA performance issue

  #9 opened 2 months ago by ashakoen
  7

• Hot Lora Replacement

  #18 opened 2 months ago by Lantianyou
  17

• [bug]UnboundLocalError: cannot access local variable 'temp_77_token_ids' where it is not associated with a value

  #23 opened 2 months ago by 81549361
  2

• TypeError: NoneType takes no arguments

  #25 opened 2 months ago by lvjin521
  4

• Load a LORA using the API

  #20 opened 2 months ago by acaladolopes
  3

• The speed of drawing is not satisfactory

  #26 opened 3 months ago by lvjin521
  4

• Why is vae decoder so slow? Can you help me?

  #27 opened 2 months ago by radish0926
  4

• LoRA loading fails if only trained on specific blocks

  #13 opened 4 months ago by fblissjr
  21

• How to save a "prequantized_flow" safetensor?

  #16 opened 4 months ago by smuelpeng
  5

• PuLID support

  #19 opened 3 months ago by 81549361
  0

• Docker image support.

  #17 opened 4 months ago by ShivamB25
  5

• [feature] Support for ControlNet from x-lab

  #6 opened 4 months ago by George0726
  3

• No issue - just a thank you!

  #4 opened 4 months ago by ashakoen
  2

• Consider adding a license to the code

  #12 opened 4 months ago by flowpoint
  4

• Error No module named 'cublas_ops'

  #5 opened 4 months ago by ankitsiliconithub
  2

• WHL files for torch-cublas-hgemm.git and ao?

  #7 opened 4 months ago by SoftologyPro
  2

• `NotImplementedError: Cannot copy out of meta tensor; no data!`

  #8 opened 4 months ago by montyanderson
  3