city96 has published GGUF versions of the Flux1.Dev model and T5 XXL text encoder, along with custom nodes to use them in ComfyUI - thought I’d try them on my M2 Mac mini, hoping for faster inference!

More Flux.1-based models! Go faster with FP8 or NF4! New LoRAs and ControlNets! There is quite a bit of interest with this model, as evidenced by the speed of community-led enhancements.

While I have many posts about SDXL, I do not use Stable Diffusion 3 at all - license concerns aside, it is simply not good, and may never get better. But just a few days ago, a new, freely available, offline model that is better than SDXL was released by the team that presented Latent Diffusion and created Stable Diffusion, Flux.1 by Black Forrest Labs!

I know it’s bad form to start off with a disclaimer: but the truth is, I do not know what I am doing. I am just testing out two new ComfyUI nodes, PerturbedAttentionGuidance and PerpNegGuider.

In my last post, I used ComfyUI-IF_AI_tools to integrate to the brxce/stable-diffusion-prompt-generator model running in Ollama. I wonder if I could use the base Mistral 7B model to help improve my uncreative prompts instead...

In my last post, I described running Mistral, a Large Language Model, locally using Ollama. To accompany that piece, I created a prompt and manually used AI to generate an image. Today, I’ll wire up a ComfyUI workflow to Ollama to do this seamlessly, thanks to ComfyUI-IF_AI_tools.

More and more AI generated images are shared as short video clips. So, here a quick test of Stable Video Diffusion - which was released back in November last year. Don’t know why I didn’t post this when I posted about AnimateDiff and the Hotshot Motion model around the same time.

Do you want to convert a 2D image into a 3D model auto-magically? On 5 March 2024, Stability AI and Tripo AI released TripoSR: Fast 3D Object Generation from Single Images that does exactly that!

Differential Diffusion is the newest method (framework) of in-painting without an in-painting model. Instead, all that is needed is a mask (map) where the lighter the area, the greater the re-painting applied.

Ever wished you could generate Stable Diffusion XL images with transparent backgrounds? Well, your wish has been answered by the smart people behind the Transparent Image Layer Diffusion using Latent Transparency paper. They have made their code and models available, and what do you know, Chenlei Hu has ported it to ComfyUI!

With the advent of techniques like Adversarial Diffusion Distillation and Latent Consistency models, A.I. image synthesis based on Stable Diffusion XL has been getting faster and faster. Here is just quick comparison of a few models at 4-steps, some of which are fine-tuned and trained for realism.

Not long ago, in a attempt to obtain Consistent portraits using IP-Adapters for SDXL, I shared a comparison between IP-Adapter-Plus-Face and IP-Adapter-FaceID. Today I’ll look at InstantID.

An update to my previous post on Stable Cascade with ComfyUI - instead of requiring four separate model files, we now only need two checkpoints, and the ComfyUI workflow is now very straightfoward!

On 12 Feb 2024, Stability.ai released Stable Cascade “research preview” (non-commercial license), and over the weekend, ComfyUI was updated to support this new model! Time to give it a go!

As a follow up to my last post regarding Consistent portraits using IP-Adapters for SDXL, this is a short comparison of the two face IP-Adapters for SDXL by h94 / xiaohu: namely, ip-adapter-plus-face_sdxl_vit-h.bin and ip-adapter-faceid_sdxl.bin.

Getting consistent character portraits generated by SDXL has been a challenge... until now! ComfyUI IPAdapter Plus (dated 30 Dec 2023) now supports both IP-Adapter and IP-Adapter-FaceID (released 4 Jan 2024)!

In the span of a couple of weeks, we got Crazy fast image generation with LCM LoRA for SDXL, which led me to ask if I could get Faster Stable Diffusion on M-series macs?. A few hours ago, Stability.ai gave us their response in the form of SDXL-Turbo... and now we go even faster!

All my recent Stable Diffusion XL experiments have been on my Windows PC instead of my M2 mac, because it has a faster Nvidia 2060 GPU with more memory. But today, I’m curious to see how much faster diffusion has gotten on a M-series mac (M2 specifically).

Time to try another ControlNet for Stable Diffusion XL - QR Code Monster v1 in ComfyUI. This ControlNet can influence SDXL such that the generated image “hides” a scan-able QR code, which at first glance, looks like a photo!

Stable Diffusion keeps improving at an astounding pace! This time, it’s the idea of distilling a model into a Latent Consistency Model (LCM) for very, very fast image generation with a quality trade-off. On 24 Oct 2023, the distilled Segmind Stable Diffusion 1B (SSD-1B) model was released, followed by a better implementation in the form of Latent Consistency LoRAs for SDXL and SDD-1B released on 9 Nov 2023.