filipstrand/mflux

Issues

• Support for Gradio WebUI

  #15 opened 3 months ago by 7enChan
  13

• Image generation speed for M4 Max Macbook Pro using mflux

  #92 opened 12 days ago by kasnol
  6

• Information Sharing on the rf_inversion Technology

  #91 opened 13 days ago by raysers
  2

• Failing tests for mlx > 0.19.0

  #93 opened 11 days ago by filipstrand
  2

• Info about Python 3.13 compatibility

  #74 opened a month ago by anthonywu
  3

• Saving to original .cache in spite of setting HF_HOME

  #89 opened 17 days ago by mokko
  2

• A Presumptuous Suggestion: Adding Localized Redraw Based on Image-to-Image (Like Inpainting)

  #88 opened 19 days ago by raysers
  4

• Make binary adhere to signals

  #79 opened 21 days ago by Morriz
  1

• If possible, could the official team consider making it compatible with ComfyUI?

  #56 opened 2 months ago by raysers
  15

• What is the relationship between steps and image2image strength?

  #86 opened 23 days ago by twalderman
  6

• Benchmark you can add - Mac Mini 2023 - Apple M2 Pro - 32gb ram - os - 15.0

  #85 opened 23 days ago by leekichko
  3

• Unable to access gated repository for FLUX.1-dev model

  #14 opened 3 months ago by 7enChan
  8

• proposal: automatic output file naming

  #80 opened a month ago by anthonywu
  3

• Performance Report - 2020 M1 Air 8GB

  #13 opened 3 months ago by mbvillaverde
  2

• Feature Request: Train lora

  #78 opened a month ago by kirel
  3

• proposals for refactoring/simplify CLI entry points

  #66 opened a month ago by anthonywu
  1

• Would it be possible to also implement PulID?

  #68 opened 2 months ago by azrahello
  1

• running lora B on saved quantized model with lora A on disk

  #76 opened a month ago by azrahello
  2

• Where folder put mlx flux dev ? for use in comfyUI ?

  #67 opened 2 months ago by stef06
  5

• Support for different LoRA formats

  #47 opened 2 months ago by filipstrand
  6

• `seed` can't be `None`

  #69 opened 2 months ago by Morriz
  3

• Can't install

  #41 opened 2 months ago by ckizer
  6

• Fetching files freeze

  #42 opened 2 months ago by filippkowalski
  2

• local path quantized model with lora not working?

  #49 opened 2 months ago by AugustRush
  2

• Project needs a license

  #54 opened 2 months ago by anthonywu
  1

• I manually downloaded the 4-bit quantized model. Can I avoid downloading the full model from Hugging Face? If I can directly use my downloaded model, what command should I use to specify it?

  #52 opened 2 months ago by raysers
  2

• TypeError: unsupported operand type(s) for |: 'type' and 'NoneType' when importing / running a command

  #53 opened 2 months ago by pelayomartinez
  1

• Question about generation speed

  #33 opened 2 months ago by TuiKiken
  1

• Tried the "official" flux-RealismLora and met with missing transformer error

  #40 opened 2 months ago by meetbryce
  16

• Installation (Macbook Pro M3)

  #44 opened 2 months ago by light-merlin-dark
  4

• 4bit model requires more than 9GB

  #36 opened 2 months ago by sudhamjayanthi
  5

• Support for Different Samplers and Schdulers?

  #45 opened 2 months ago by andyw-0612
  1

• Error installing 0.2.0

  #38 opened 2 months ago by heavysixer
  2

• controlnet support ?

  #43 opened 2 months ago by AugustRush
  0

• `Unable to load kernel winograd_conv_2d_weight_transform_bfloat16_bc32`

  #37 opened 2 months ago by marckohlbrugge
  2

• Where do I need to place the Lora file for it to work as shown in the examples?

  #39 opened 2 months ago by meetbryce
  2

• How use other flux base model ?

  #34 opened 2 months ago by douenergy
  1

• Parameter for limiting CPU usage

  #17 opened 2 months ago by kennell
  1

• Load model from drive, note hf cache

  #19 opened 2 months ago by stefanvarunix
  3

• Feature Request: Show Prompt Metadata in Get Info

  #26 opened 2 months ago by vincyb
  2

• Is there some max prompts length limit?

  #28 opened 3 months ago by smoothdvd
  1

• Why GPU usage is low?

  #25 opened 3 months ago by smoothdvd
  2

• 4-bit

  #18 opened 3 months ago by okaneyo
  3

• Can't install modules mlx and either torch or numpy via pip

  #23 opened 3 months ago by t-beaver
  1

• Where does it download the model ?

  #20 opened 3 months ago by dostarora97
  1

• Performance report - 2023 M2 Max 96GB

  #6 opened 3 months ago by explorigin
  10

• Performance report - 2023 M3 Pro 36GB

  #11 opened 3 months ago by kush-gupt
  2

• `--output` not being used

  #12 opened 3 months ago by rafrafek
  1

• Great Work! and works like a charm

  #8 opened 3 months ago by vincyb
  1

• Performance report - 2021 M1 Pro 16GB

  #7 opened 3 months ago by qw-in
  1