Andrew

Forum Replies Created

Viewing 15 posts - 16 through 30 (of 233 total)
  • Author
    Posts
  • in reply to: Runtime errors, again… Help #17403
    AndrewAndrew
    Keymaster

      Haven’t encountered this before but you can try:

      1. Use SDP Attention “–opt-sdp-attention” argument.

      2. Forge uses a different backend. Try if you see the same error.

      in reply to: ComfyUI Checkpoint Lottery with Image generation. #17402
      AndrewAndrew
      Keymaster

        Not easy to coment on the specifics without an example, but you can:

        1. Test the same prompt on the SDXL 1.0 base model. It is well trained and can give you an ideal whether the juggernaut model has training issues.
        2. Whether it is due to the association effect https://stable-diffusion-art.com/courses/stable-diffusion-level-2/lessons/association-effect/
        3. Use negative prompt to supresss anything and settings you don’t want to see.
        in reply to: Add direct download to Colab notebooks #17354
        AndrewAndrew
        Keymaster

          Yes it is possible but I want to keep the UI simple.

          If there’s enough interest, I can create another notebook for downloading models to google drive.

          in reply to: Forge Notebook Not Working? #17352
          AndrewAndrew
          Keymaster

            Hi, both the A1111 and Forge notebook are now fixed.

            in reply to: Fix Flux Plastic Skin? #17351
            AndrewAndrew
            Keymaster

              Interesting. An option is to combine them – use flux to generate the initial image followed by SDXL with img2img at a low denoising strength.

              in reply to: Fix Flux Plastic Skin? #17228
              AndrewAndrew
              Keymaster

                Hi, can you post a Flux image that looks plastic and one from Juggernaut that you think is good?

                I need to see what you are looking for before giving advice.

                in reply to: Moder and Clip Loader #17208
                AndrewAndrew
                Keymaster

                  LoRA should be the go-to method for modifying models. A LoRA can also modify CLIP, but the main effect is in modifying the diffusion model.

                  I haven’t done a comparison, but sampling is a pretty standard process. I don’t think they would do anything different.

                  in reply to: Moder and Clip Loader #17196
                  AndrewAndrew
                  Keymaster

                    Hi, we need three models to use a diffusion model like Stable Diffusion and Flux

                    • Diffusion model – for denoising during sampling.
                    • VAE – for converting the images between pixel and latent spaces.
                    • CLIP – for encoding to text prompt for conditioning during sampling.

                    Some checkpoints include all three in a single checkpoint file. Even if others don’t, the Load Checkpoint node uses the default VAE and CLIP models.

                    In addition to the original VAE, there are improved or finetuned versions, although they are rare. You can use the Load VAE node to specify the VAE you want. Typically, the difference is minimal.

                    Likewise, you can load the CLIP models directly using a node. Some models, like Flux, use two text encoders. Using the Dual CLIP loader allows you to put different text prompts to different encoders. Some people swear to see a difference between the two, but this remains an under-explored area.

                     

                     

                     

                    in reply to: Moder and Clip Loader #17195
                    AndrewAndrew
                    Keymaster

                      Hi, we need three models to use a diffusion model like Stable Diffusion and Flux

                      Diffusion model – for denoising during sampling.
                      VAE – for converting the images between pixel and latent spaces.
                      CLIP – for encoding to text prompt for conditioning during sampling.
                      Some checkpoints include all three in a single checkpoint file. Even if others don’t, the Load Checkpoint node uses the default VAE and CLIP models.

                      In addition to the original VAE, there are improved or finetuned versions, although they are rare. You can use the Load VAE node to specify the VAE you want. Typically, the difference is minimal.

                      Likewise, you can load the CLIP models directly using a node. Some models, like Flux, use two text encoders. Using the Dual CLIP loader allows you to put different text prompts to different encoders. Some people swear to see a difference between the two, but this remains an under-explored area.

                       

                       

                       

                      in reply to: AnimatDiff error in Comfy #17069
                      AndrewAndrew
                      Keymaster

                        It’s great to see you figured it out. It’s a warning, so nothing to worry about.

                        in reply to: AnimatDiff error in Comfy #17057
                        AndrewAndrew
                        Keymaster

                          Mmm… The only things I can think of are:

                          • Use a T4 instance (that’s what I used)
                          • Remove or rename the custom_nodes folder in AI_PICS > ComfyUI
                          in reply to: AnimatDiff error in Comfy #17055
                          AndrewAndrew
                          Keymaster

                            Hi David, I just tested the workflow on Colab, and it works correctly. I used the dreamshaper model and the motion model in the workflow. Did you change any settings?

                            in reply to: ControlNet modes not loading in ComfyUI #16959
                            AndrewAndrew
                            Keymaster

                              Good find! I have fixed the issue.

                              in reply to: Flux Gym for training Loras models #16948
                              AndrewAndrew
                              Keymaster

                                I haven’t used it but it uses Kohya in the backend, so it should work.

                                in reply to: ComfyUI beginners course resize images dead link #16934
                                AndrewAndrew
                                Keymaster

                                  Yes, I am in the process of writing the lessons. Should be up within a week.

                                Viewing 15 posts - 16 through 30 (of 233 total)