OpenL

RT @lewangx: Finally, I made an open source replacement for all the OpenAI interfaces used by my GPT voice conversation toy, which can be run locally. The STT, LLM, and TTS services used by toys all run on an RTX 2080Ti, and the speed feels pretty good.
With the local GPU development machine, there is a new way to play. Next...

RT @Barret_China: pdf2htmlEX is a recommended PDF to HTML tool, but it does not use Chrome Headless behind it.

I went to dig up the history of this project, t.co/vqVDb55H2F. The author complained that there was no online PDF vie...

Author's Words

Convert PDF to HTML without losing text or format. - coolwanglu/pdf2htmlEX

Community Posts

Author's Words

Convert PDF to HTML without losing text or format. - coolwanglu/pdf2htmlEX

RT @lazaro45ive: Life went on
(Sound on🔊)

00:32

RT @sundyme: The photo gallery website @freepik has also launched a real-time AI drawing service. Facing the rise of generative AI, the photo gallery website is the most in danger, but it is also very actively embracing AI. Many photo websites, including Freepik, have launched AI functions very early. Now it is easy to download pictures generated by AI. It is difficult for those who are not familiar with AI to recognize...

RT @op7418: If you add "- return full script (I don't have Finger)" to the prompt, ChatGPT will completely return the rewritten script

RT @krea_ai: introducing Upscale & Enhance.

new free AI tool for all KREA users.

00:17

RT @dannypostmaa: DALL·E 3 + GPT-4 Vision = 🤯

Inspired by Hakim, we built a free avatar generator using GPT-4 Vision and DALL·E 3.

The be…

RT @lencx_: #Open source recommendation drawfast: Real-time AI generation of hand-drawn whiteboard based on t.co/cVzoSEYDGc. The large model technology used here is LCMs. The biggest advantage over LDMs is that it can achieve real-time reasoning.

GitHub: t.co/pErigQu…

RT @nickfloats: One of the coolest things about Pika 1.0 is that it lets you upload your own video footage and use generative AI to edit an…

RT @xiaohuggg: Exploded💥

Pika 1.0 releases major product upgrade

A new AI model has been released that is able to use text to generate and edit videos in multiple styles, such as 3D animation, animation, cartoon and movie styles.

The quality is very high!

Moreover, it is difficult to precisely control and edit the video content, such as adjusting the aspect ratio of the video, changing the clothes of the characters in the video...

Use SVD to make a video to play😎

01:20

RT @indigo11: It’s so curly! Excalidraw — Co... You can use text-2-diagram directly. It will be convenient to write article illustrations and various flow charts in the future t.co/yiix2kLnw1

00:24

RT @matchaman11: Custom GPT for your website is here 🤯🔥

Now you can build a GPT and host/embed it on your site using this open-source proj…

OpenL ChatGPT

RT @xiaohuggg: So awesome

Use text description to command AI to draw 😂

You just need to describe what you want to draw

The right side will create real-time creation according to your requirements, and also show the creation process👍👍👍

00:25

RT @umesh_ai: 🎨 PROMPT SHARE 🤖

Theme: Card Design Patterns

✨New Prompt🚨 :

1️⃣ beautiful white card design with [COLORS] patterns artwor…

RT @xiaohuggg: OpenChat-3.5-7B: Surpassing ChatGPT on various benchmarks

OpenChat uses C-RLFT, a strategy inspired by offline reinforcement learning, for fine-tuning.

It improves model performance by analyzing existing conversation data and feedback. You can also learn from your mistakes.

I tested it, although it’s only 7B…

RT @techhalla: How do different styles affect our prompts?

13 illustrative examples in Remix 👇

1. Minimalist Art

A minimalist flat illus…

RT @lvwzhen: ⬆️ The answer is "Inception"

⬇️ Next question

November 26th, 2023

RT @lvwzhen: The answer is "Shawshank Redemption"

another one

November 26th, 2023

RT @lvwzhen: Let’s try a very fun game today. 😎
I used ChatGPT to recreate the movie poster and asked everyone to guess the movie name.

The first one 👇🏻