Pinned
Shocked! Llama3-V project from a Stanford team plagiarized a lot from MiniCPM-Llama3-V 2.5! its code is a reformatting of MiniCPM-Llama3-V 2.5, and the model's behavior is highly similar to a noised version of MiniCPM-Llama3-V 2.5 checkpoint. Evidence: github.com/OpenBMB/MiniCP…- I have found how to train FLUX.1 lora，It seems like need 48G VRAM at least，here is a guidelines for training. github.com/bghira/SimpleT… @bdsqlsz @araminta_k @AIWarper @ZHOZHO672070 @camenduru @cubiq
- CatVTON which is a latest VTON project Comfyui node has released github.com/pzc163/Comfyui… CatVTON is a simple and efficient virtual try-on diffusion model, below for more catVTON info - code: github.com/Zheng-Chong/Ca… - HF: huggingface.co/zhengchong/Cat… - paper: arxiv.org/abs/2407.15886
- My first flux-Dev Lora model training was completed successfully. This is a graphic illustration style lora. I used 78 images some of them from @araminta_k .I will upload the lora model to huggingface later prompt: a girl walking on the street,surrounded by a group of pigeons
- flux lora has been released on:
- Good news! I spend several days on fine-tuning a Prompt generation VLM model.The fine-tuned model is trained on a 3000 samples midjourney prompt dataset . The model can generate short prompts and long prompts for images with natural language style.
- Replying to @yangzhizheng1 @OpenBMB and 7 othersLlama3-V's model structure and config.json file are identical to miniCPM-llama3-v2.5.I don't know its parameters.But guess what you get if you add Gaussian noise to MiniCPM-Llama3-V 2.5's checkpoint? Let's have a look！
- Replying to @siddrrsh and @AkshGarg03It's obviously not wise to blame one person for the thefts of all three.
- Replying to @yangzhizheng1Fact 1: The llama3-V project uses exactly the same model structure and code as the minicom-llama 3-v 2.5 project Llama3-V has exactly the same model structure and config file as MiniCPM-Llama3-V 2.5, with only the difference in variable names.
- Replying to @yangzhizheng1 @OpenBMB and 7 othersThat's crazy! You can actually get a new checkpoint, emm, so let's give this new checkpoint a new name and call it llama3-V, doesn't that sound great? At least the hash will be completely different from miniCPM-llama3-V2.5, right?🤣🤣🤣🤣🤣
