CarperAI / DRLX
Diffusion Reinforcement Learning Library
☆173Updated 8 months ago
Related projects ⓘ
Alternatives and complementary repositories for DRLX
- Huggingface-compatible SDXL Unet implementation that is readily hackable☆398Updated last year
- This is the official repo for the paper "Zeroth-Order Optimization Meets Human Feedback: Provable Learning via Ranking Oracles", Tang et…☆195Updated last year
- Faster generation with text-to-image diffusion models.☆192Updated last month
- This repository implements the idea of "caption upsampling" from DALL-E 3 with Zephyr-7B and gathers results with SDXL.☆153Updated last year
- Simple large-scale training of stable diffusion with multi-node support.☆125Updated last year
- Better Aligning Text-to-Image Models with Human Preference. ICCV 2023☆265Updated last year
- Iterable datapipelines for pytorch training.☆81Updated 2 months ago
- ☆110Updated 2 years ago
- ☆317Updated last month
- AlignProp uses direct reward backpropogation for the alignment of large-scale text-to-image diffusion models. Our method is 25x more samp…☆241Updated last week
- Reproduction of DDPO paper (RLHF for diffusion)☆68Updated last year
- Open reproduction of MUSE for fast text2image generation.☆331Updated 5 months ago
- Repo is required for the code of our research paper on micro-budget training of large scale diffusion model.☆153Updated 3 months ago
- ☆164Updated 2 years ago
- ☆119Updated last month
- Implementation of HyperDreamBooth: HyperNetworks for Fast Personalization of Text-to-Image Models☆165Updated last year
- ☆437Updated 9 months ago
- ☆406Updated 7 months ago
- Implementation of "SVDiff: Compact Parameter Space for Diffusion Fine-Tuning"☆369Updated 9 months ago
- Code for instruction-tuning Stable Diffusion.☆210Updated 8 months ago
- ☆85Updated last year
- Implementation of Key-Locked Rank One Editing, from Nvidia AI☆230Updated last year
- T-GATE: Temporally Gating Attention to Accelerate Diffusion Model for Free!☆358Updated 2 months ago
- Official PyTorch codes for the paper: "ViCo: Detail-Preserving Visual Condition for Personalized Text-to-Image Generation"☆238Updated 7 months ago
- ☆71Updated last year
- Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis☆395Updated 5 months ago
- Implementation of Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Models☆321Updated last year
- Code for "Diffusion Model Alignment Using Direct Preference Optimization"☆261Updated 10 months ago
- Erasing Concepts from Diffusion Models☆549Updated last week
- SSD-1B, an open-source text-to-image model, outperforming previous versions by being 50% smaller and 60% faster than SDXL.☆166Updated 6 months ago