martyn / safetensors-merge-supermarioLinks
Merge safetensor files using the technique described in "Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch"
☆80Updated 10 months ago
Alternatives and similar repositories for safetensors-merge-supermario
Users that are interested in safetensors-merge-supermario are comparing it to the libraries listed below
Sorting:
- ☆128Updated 11 months ago
- Scripts for use with LongCLIP, including fine-tuning Long-CLIP☆62Updated 6 months ago
- Unofficial implementation. Stable diffusion model trained by AI Feedback-Based Self-Training Direct Preference Optimization.☆65Updated last year
- Greatly increase the diversity of your generated images in Automatic1111 WebUI through Condition-Annealed Sampling.☆108Updated last year
- A comprehensive codebase for training and finetuning Image <> Latent models.☆44Updated 6 months ago
- ☆54Updated 2 years ago
- ☆37Updated last year
- (IA)^3 for Stable Diffusion☆35Updated 2 years ago
- Model code for inferencing T5☆66Updated 6 months ago
- This repository implements the idea of "caption upsampling" from DALL-E 3 with Zephyr-7B and gathers results with SDXL.☆156Updated last year
- ComfyUI powertools for SD1.5 and SDXL model merging☆93Updated 5 months ago
- Official codebase for Margin-aware Preference Optimization for Aligning Diffusion Models without Reference (MaPO).☆81Updated last year
- Set of Utilities I Have Coded to Help Me Train RPGv6 on Flux1☆82Updated last year
- ☆126Updated 6 months ago
- Diffusers Implementation of Controlling Text-to-Image Diffusion by Orthogonal Finetuning☆35Updated last year
- finetune your florence2 model easy☆20Updated last year
- ☆42Updated last year
- Diffusion Reinforcement Learning Library☆191Updated last year
- Uses DARE to merge LoRA stacks as a ComfyUI node☆34Updated last year
- ☆435Updated last year
- SliderSpace: Decomposing the Visual Capabilities of Diffusion Models☆100Updated 6 months ago
- Sparse Autoencoders (SAE) vs CLIP fine-tuning fun.☆16Updated 8 months ago
- SSD-1B, an open-source text-to-image model, outperforming previous versions by being 50% smaller and 60% faster than SDXL.☆177Updated last year
- ☆61Updated last year
- Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models☆50Updated 2 years ago
- ☆50Updated last year
- Vision Transformers Needs Registers. And Gated MLPs. And +20M params. Tiny modality gap ensues!☆47Updated 3 months ago
- This extension provides inference-time optimization techniques to enhance diffusion-based image generation quality through random search …☆22Updated 6 months ago
- Modern Stable Diffusion models family - Fluently☆32Updated last year
- Executable State Dict Recipes☆71Updated this week