Oddbean new post about | logout
 # Nova IA de Imagens de Código Aberto

A IA OmniGen permite geração de edições ou novas imagens a partir de uma imagem inicial, no modelo 'image to image'.

https://aiomnigen.com/

https://github.com/VectorSpaceLab/OmniGen 
 nostr:nevent1qqsfwus69j3ngu5a4uzm2g5uqczg8plqxzy299y8ze90x7zaglfspxcpzpmhxue69uhkummnw3ezumt0d5hsyg99aez8269zxu3zd4j8qya92fg7437ax745pqz22ys6v08zef65qypsgqqqqqqsm9dl3j 
 O site ainda está bem em fase beta.
A ideia de merge duas imagens é bem legal, mas estou testando e está demorando muito. 
 A depender do seu uso, tem esses outros:

nostr:nevent1qqsxdpwadkswdrc602m6qdhyq7n33lf3wpjtdjq2adkw4y3h38mjcrqpr9mhxue69uhkxmmzwfskvatdvyhxxmmd9aex2mrp0ypzpf0wg36k3g3hygndv3cp8f2j284v0hfh4dqgqjj3yxnreck2w4qpqvzqqqqqqysn06gs 
 Se não tiver no Amethyst, a ultima edição é essa:


# AI Models and Tools
___

#### Open Source
___

#### Models
- [LLAMA](https://www.llama.com/) - Large Language Model Meta AI - Text AI [[source]](https://github.com/meta-llama/llama-models)
- [MistralAI](https://mistral.ai/) - Text AI [[info]](https://mistral.ai/company/) [[source]](https://github.com/mistralai/mistral-inference)

- [Stable Diffusion](https://stability.ai/stable-image) -  Text to image [[source]](https://github.com/Stability-AI/stablediffusion)
- [Pixart-Alpha](https://pixart-alpha.github.io) - Text to image [[source]](https://github.com/PixArt-alpha/PixArt-alpha)
- [OmniGen](https://aiomnigen.com/) - Pompt, image or subject to image [[source]](https://github.com/VectorSpaceLab/OmniGen)
- [Pigallery](https://bpatrik.github.io/pigallery2) - Self-Hosted AI Image Generator [[source]](https://github.com/bpatrik/pigallery2)

- [CogVideo](https://github.com/THUDM/CogVideo) - Text and image to video generation [[info]](https://zhipu-ai.feishu.cn/wiki/DHCjw1TrJiTyeukfc9RceoSRnCh) [[source]](https://github.com/THUDM/CogVideo) [[test]](https://huggingface.co/spaces/THUDM/CogVideoX-5B-Space)

___
#### Tools
- [HuggingFace](https://github.com/huggingface/transformers) - Test and collaborate on models, datasets and apps. [[source]](https://github.com/huggingface)
- [DuckDuckGo AI Chat](https://duckduckgo.com/duckduckgo-help-pages/aichat/) - Famous AIs without Login [[source]](https://github.com/duckduckgo/duckduckgo-help-pages/commits/master/_docs/aichat/aichat-index.md)
- [Ollama](https://ollama.com/) - Run LLMs Locally [[source]](https://github.com/ollama/ollama)
- [DreamStudio](https://dreamstudio.com) - Stable Diffusion’s Web App Tool [[info]](https://dreamstudio.com/about/) [[source]](https://github.com/Stability-AI/generative-models)
- [Prompt Gallery](https://www.promptgallery.app/) - AI images with their prompts [[source]](https://github.com/SchmitzAndrew/prompt-gallery)
___

#### Closed Source
___

#### Models

- [ChatGPT](https://openai.com/chatgpt/overview/)
- [Claude](https://www.anthropic.com/claude)
- [Gemini](https://gemini.google.com)
- [Copilot](https://www.microsoft.com/microsoft-copilot/personal-ai-assistant)
- [Midjourney](https://www.midjourney.com/home)
- [Minimax](https://hailuoai.com/) - Realistic videos (prompts only in chinese)
___

#### Tools

- [Meshy](https://www.meshy.ai/) - 3D Generative AI Toolbox [[info]](https://docs.meshy.ai/)

___

Other index: [Amazing AI](https://github.com/openbestof/awesome-ai?tab=readme-ov-file)