Copied 12/09/2022 from docs.google.com/spreadsheets
Text to Image | ||
Name | Description | Price |
Research papers from google lab on their text to image model. | ||
OpenAI text to image model with outpainting feature. | ||
Free Dall-e 2 and Stablediffusion credit (limited) and search engine for AI-generated images (Dalle-2, Midjourney, Stable Diffusion) with prompts. | ||
StabilityAI text to image model with outpainting feature. | ||
Gradio based project of Stable Diffusion WebUI with a lot of features. | ||
Dataset of prompts, synthetic AI generated images, and aesthetic ratings. | ||
Text to image model based on Stable Diffusion. | ||
Midjourney is currently best text to image model. | Free Trial, paid access | |
MetaAI research papers on their text to image model (Unrelesed). | ||
Website to find similar AI generated image. | ||
Microsoft text to image model. | Free to use | |
New chinese text to image model created by Baidu. The model has great coherence with a lot less image dataset than competitors. | Free demo | |
DeepAI text to image model based on Stable Diffusion. | Free to use | |
Site is down. | Free | |
Old text to image model created by community originaly named "Dall-e mini". | Free | |
DALL·E Flow generates HD images from text prompts by first leveraging DALL·E-Mega, GLID-3 XL, and Stable Diffusion to generate image candidates, and then calling CLIP-as-service to rank the candidates. The preferred candidate is then fed to GLID-3 XL for diffusion, and finally upscaled to 1024x1024 via SwinIR. | Free | |
Text to image model created by Wombo. | Free & Paid | |
Erlich is the text to image latent diffusion model from CompVis (with additions from glid-3-xl) finetuned on a dataset collected from LAION-5B. | ||
Stable Diffusion fine tuned on Pokemon dataset. | ||
Latent Diffusion is a text-to-image model created by CompVis, trained on the LAION-400M dataset. | ||
A 1.4B parameter text-to-image model from CompVis, finetuned on CLIP text embeds and curated data. | Free & paid | |
Instagram like website with AI generated images. | Free & paid | |
Text-to-image model created by JinaAI tuned to produced best quality images in style of oil painting. | Free & paid | |
Free & unlimited access to Stable Diffusion v1.5 without filters. | ||
Another text to image free tokens. | Free & paid | |
Tool to creates photorealistic images from segmentation maps, which are labeled sketches that depict the layout of a scene. Artists can use text, a paintbrush and paint bucket tools, or both methods to design their own landscapes. | Free to use | |
Anime/Furry Generation model. | Paid | |
Art generation website with interesting algorithm for create animation, image from sketch and more. | ||
Website powered by Stable Diffusion v1.5, with finetuned autoencoder (f8-ft-MSE). | ||
Stable Diffusion and Disco Diffusion models. App allows you to generate images from a simple drawing. | ||
CompVis and Stability AI model demo for sketch. | ||
Art generator app uses two AI models to generate art. First one being, Altair, which uses VQGAN-CLIP model to render the artwork creations. The second is, Orion, which uses CLIP-Guided Diffusion to create artworks and imageries. | ||
Website develop custom neural network solutions and models for image generation, upscaling and more. | ||
Text Generation Models | ||
Name | Description | Price |
Currently the most powerful text model available on the market. | ||
The world largest open-science, open-access multilingual large language model, with 176 billion parameters, allow text generation in 46 languages. | ||
Powerful 178B-parameter language text model. The $90 in free credits allows to test fine-tuning models. | ||
Powerful NLP model to generate text, embed question and classify stuff. Free | ||
Implementation of model & data-parallel autoregressive language models, utilizing Mesh Tensorflow for distributed computation on TPUs. | ||
Megatron 11b? | ||
T5 is an encoder-decoder model and converts all NLP problems into a text-to-text format. | ||
Text to Video | ||
Name | Description | Price |
Research papers about text-to-video model. A lot better than first version of Meta model. | ||
Research papers about text-to-video model. | ||
Large-scale Pretraining for Text-to-Video Generation via Transformers. | ||
Text to Code | ||
Name | Description | Price |
Natural language-to-code system based on GPT-3, helps turn simple English instructions into code. | ||
GitHub Copilot is an AI pair programmer that offers autocomplete-style suggestions as you code. | ||
Open Source competitor of Codex model. Trained on larger dataset than Codex. | ||
Tabnine suggests code completions that align with previous coding patterns. The plug-in is available on many more IDEs and text editors than copilot. | ||
AI pair programming tool similar to GitHub Copilot. | ||
Free AI Coding Assistant and Code Auto-Complete Plugin. | ||
Text to 3D Model | ||
Name | Description | Price |
Google lab research papers about text-to-3d model. Community implementation can be found here. | ||
Model turns couple images of an object into a smoothly transitioning animation. | ||
Audio Generation | ||
Name | Description | Price |
Text-to-audio model that can create different sounds for example "whistling while wind blowing". | ||
Deep neural network that can generate 4-minute musical compositions with 10 different instruments, and can combine styles from country to Mozart to the Beatles. | ||
Website with AI model to create high quality royalty-free music. | ||
Another website for generating music. | ||
This AI model can create music from text or video. | ||
Neural net that generates music, model is created by OpenAI. | ||
Another AI for making music. I don't test it yet. | ||
Powerful audio and video editing AI tool. | ||
Open source neural net that approaches human level robustness and accuracy on English speech recognition. Model is multilingual but work best on english language. | ||
Demo of General Speech Restoration With Neural Vocoder. | ||
The Artificial Intelligence composing emotional soundtrack music. | ||
Text to Speech | ||
Name | Description | Price |
Create different characters voice from text. | ||
Another model to create speech from text. Website allow to create own fake voice. | ||
You can choose from over 2,000 voice cloning options to let you imitate anyone from Donald Trump to Sir Mix-A-Lot. | ||
Upscalers & Restorers | ||
Name | Description | Price |
Image Restoration Using Swin Transformer. | ||
Powerful general upscaler made by Topaz lab. | ||
The best model for enhancing and upscale old or poor quality face photos. | ||
A professional Chinese AI based video enhancing and upscaling tool for video quality enhancement and resolution upscaling. | ||
A professional Chinese AI based photo enhancing and upscaling tool for photo quality enhancement and resolution upscaling. | ||
A professional Chinese AI based face enhancing and upscaling tool for quality enhancement and resolution upscaling. | ||
open-source AI image upscaler trained with pure synthetic data. It extends ESRGAN (Enhanced Super-Resolution Generative Adversarial Networks). | ||
Website develop custom neural network solutions and models for image generation, upscaling and more. | ||
Realtime Voice Changing | ||
Name | Description | Price |
Powerful AI tool to change your voice. | ||
Realistic voice changer. | ||
Most popular AI voice changer. | ||
Audio software and plug-ins for mixing, mastering, restoration, and more. | ||
Video Avatars | ||
Name | Description | Price |
AI avatar with best quality. You can add versions of yourself with one photo. | ||
Also good quality AI avatar. | ||
Low & Medium Quality Avatars. | ||
Purchase Comes With Unlimited Videos Of 1 Minute or Less, Medium Quality Avatars. | ||
Chat-bots | ||
Name | Description | Price |
You can have conversations with various famous people and characters from books or movies. | ||
Useful general AI Websites | ||
Name | Description | Price |
Website let run open-source models with a cloud API. Not open for everyone to create yet. | ||
Hugging Face is open-source and biggest platform provider of machine learning technologies. Here you can find the largest number of AI models and datasets, as well as demo versions of models. | ||
A website that helps create good prompts for text to image models. | ||
A website that helps create good prompts for text to image models. | ||
AI art made by the community 23 models: text, image video and audio prompts! Diffusions, GANs, presets and more. | ||
Free Dall-e 2 and Stablediffusion credit (limited) and search engine for AI-generated images (Dalle-2, Midjourney, Stable Diffusion) with prompts. | ||
The Stable Diffusion images search engine. | ||
This is a model from the MagicPrompt series of models, which are GPT-2 models intended to generate prompt texts for imaging AIs, in this case: Stable Diffusion. | ||
Collection of text-to-image models demos. | ||
Collection of text-to-video models demos. |