Все модели

Flux fine tuned on the 14 black paintings by Goya

one-shot-talking-face-replicate

Segment Anything MASK

Pemisah suara temannya musisi

Create cool looking gold metal things!

Generates pop art, comic book, style images

Qwen-7B is the 7B-parameter version of the large language model series, Qwen (abbr. Tongyi Qianwen), proposed by Aibaba Cloud. Qwen-7B`is a Transformer-based large language model, which is pretrained on a large volume of data, including web texts, books,

ASR model, created by Nvidia, with word-level timestamps available. Supports .wav inputs, or m3u8 urls, with a start and end time (to only process a section of the m3u8).

Cost-optimized MMAudio V2 (T4 GPU): Add sound to video using this version running on T4 hardware for lower cost. Synthesizes high-quality audio from video content.

This project uses the Segment Anything 2 (SAM2) model to remove backgrounds from videos.

Good old controlnet + inpaint + lora

Generates logograms in the style of the alien language from the 2016 sci-fi film "Arrival".

text-to-image generation

Retro style Flux Lora

Monke meme maker

CosyVoice2-0.5B-Scalable Streaming Speech Synthesis with Large Language Models

A Vision-Language Model with An Ensemble of Experts

Synthetic image detection and model identification

Test out fast inference with ExLlama and 4bit quantization!

An experiment: a fine-tuned FLUX model for UI design generation

The Yi series models are large language models trained from scratch by developers at 01.AI.

This fine-tuned SDXL model is trained with Overwatch heroes images

my own trained set

Pushing the Limits of Mathematical Reasoning in Open Language Models - Instruct model

The smartest, fastest, most useful model yet, with built-in thinking that puts expert-level intelligence in everyone’s hands

Flux lora, fine tuned on NEO-1X robot, use "NEO1X" to trigger image generation

Open-source Distilled Stable Diffusion 100% speedup

Convert a set of frames to a video

SDXL fine-tuned on energy drink designs

Detects objects in an image

Generate Renaissance paintings

YOLOv10: Real-Time End-to-End Object Detection

PyTorch version of Lightweight OpenPose as introduced in "Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose"

An Open Source text-to-speech system built by inverting Whisper

Flux LoRA, trigger style with "in the style of TOK", Photographic images with a beautiful glow

Latent Consistency Model (LCM): SSD-1B, is a LCM distilled version that reduces the number of inference steps needed to only 2 - 8 steps

An SDXL model Finetuned for style on a bespoke dataset of Stylized Illustrations