Lists (3)
Stars
Speech-to-video inference: an avatar video generated from a reference frame and a speech track
llama.cpp fork that abliterates at inference
Clip chaining for Kandinsky 6.0 in ComfyUI, also with an option to set custom audio
Run GLM-5.3-Flash (321B), DeepSeek-V4.1-Flash (748B) and MiniMax-M3.1-Flash (428B) on your own GPUs - one engine, one dashboard, OpenAI- and Anthropic-compatible API.
Kandinsky 6.0 Video SR: Models for Video Super-Resolution
Kandinsky 6.0 Video: Foundation Models for Synchronized Video and Audio Generation
FLUX 3 Action: open weights 7B world action model from Black Forest Labs. Train, fine-tune and run action prediction for robots (DROID, SO-101), simulators and games.
Call LLM and VLM in a simple way using the OpenAI API standard within ComfyUI
Train and chat with a frozen language model coupled to the full retained MaleCNS fly connectome.
Story Creation AI Agent for novel, scripts, translation, interactive games, and IP content
Optuna-optimized "abliteration" (directional ablation) directly on quantized GGUF models via llama.cpp. The base GGUF is never modified — trials run as hot-swapped LoRA adapters, and the result exp…
Kandinsky WM 1.0 — a family of models for Physical AI. Image-to-video generation for autonomous driving, robotics & general physics.
A SillyTavern extension that adds a "Delete All But This Swipe" option to the message deletion prompt, allowing you to discard every unused swipe on a message in one action while keeping the one cu…
KVAE-Audio: a continuous full-band audio waveform autoencoder
NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.
The most powerful local music generation model that outperforms almost all commercial alternatives, supporting Mac, AMD, Intel, and CUDA devices.
[CVPR 2024] RoMa: Robust Dense Feature Matching; RoMa is the robust dense feature matcher capable of estimating pixel-dense warps and reliable certainties for almost any image pair.
DisCa: Accelerating Video Diffusion Transformers with Distillation-Compatible Learnable Feature Caching
Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.
Official Implementation of OmniWeaving: Towards Unified Video Generation with Free-form Composition and Reasoning
A ComfyUI plugin that wraps [See-through](https://github.com/shitagaki-lab/see-through) — an AI system that decomposes a single anime illustration into manipulatable 2.5D layer-decomposed models wi…
"Single-image Layer Decomposition for Anime Characters" (SIGGRAPH 2026 Conference Paper)
Tools for merging pretrained large language models.
What's your LLM's personality? Measure warmth, patience, confidence and 5 more axes from hidden states.