News for hackers. Stuff that matters.

bingbing123

bingbing123 on Hacker News

Re:Nano Banana 2 and Seedream 5.0 Lite image model on NewYouGo on Friday February 27, 2026 @11:20AM (#47182327)
newyougo have a lot of models include image model , video model, music model etc...

Show HN: NewYouGo – A Fast and Free AI Image and Video Generator

Posted by bingbing123 on Wednesday January 28, 2026 @08:58AM
from the show-hn dept.

https://newyougo.com/

I’d like to share an AI tool called NewYouGo. It’s a web-based AI image and video generation platform that allows users to generate high-quality images from text prompts, with additional support for image editing and creative workflows.NewYouGo integrates several powerful open-source models (such as the Klein series and multi-angle models), and offers flexible controls for style, aspect ratio,...

( Read More... | no comments | on HN )
Re:Flux2kle.in Fast and Free Image Generator on Friday January 23, 2026 @01:02PM (#46735580)
Flux.2 klein is deployed and optimized by flux2kle.in using open-source AI models for fast and high-quality image creation.
Re:Show HN: Tikpal- Your AI Voice Partner – Focus, Flow, Forge on Friday December 26, 2025 @04:42AM (#46390660)
Yeah, we can talk about it.
Re:Show HN: Tikpal- Your AI Voice Partner – Focus, Flow, Forge on Friday December 26, 2025 @04:41AM (#46390655)
OK, our team members have already contacted you.

Show HN: Random Word Generator–a simple tool for instant writing&naming prompts

Posted by bingbing123 on Tuesday December 23, 2025 @12:43PM
from the show-hn dept.

https://randomwordgeneratorapp.top/en

I built a simple Random Word Generator that instantly generates nouns, verbs, adjectives, or mixed words.The original motivation was pretty practical: - I often need placeholder words for UI prototypes - quick prompts for writing or brainstorming - random inputs to test small tools or scriptsNo login, no tracking, just click and get words.It’s free and...

( Read More... | no comments | on HN )

Show HN: Tikpal- Your AI Voice Partner – Focus, Flow, Forge

Posted by bingbing123 on Sunday December 07, 2025 @05:27AM
from the show-hn dept.

https://tikpal.ai

We’re building Tikpal, an AI voice productivity tool based on a simple principle: Human creativity should remain the core engine. AI should be an accelerator, not the protagonist.The goal is to reduce screen dependency and cognitive fragmentation, and let people work in a more natural “voice-first” flow. Instead of clicking through interfaces and context-switching between apps, you talk to...

( Read More... | 4 comments | on HN )

Show HN: Gempix2 – A Cheap, Fast AI Image Generation API for Developers

Posted by bingbing123 on Friday November 21, 2025 @06:23AM
from the show-hn dept.

https://gempix2.us/

I built Gempix2, a minimal AI image generation service with a cheap, straightforward API for developers who don’t want to pay OpenAI / Midjourney level pricing.Why I built thisA lot of devs just need a reliable and inexpensive API to generate: • product images • marketing visuals • anime / portrait styles • automation workflow assets (Zapier / n8n / Python scripts)But most services are...

( Read More... | no comments | on HN )
Re:Nano Banana 2 – New 4K-Level AI Image Model Just Dropped on Thursday November 20, 2025 @01:07PM (#45995644)
Gempix2 (also called Nano Banana 2) is a new image generation model with strong 4K-level output, fast rendering, and surprisingly good multi-image fusion. It also maintains character consistency across scenes and handles detailed prompts well. Early access demo: gempix2.us

Show HN: Free Online Video Editor Powered by FFmpeg (Runs in the Browser)

Posted by bingbing123 on Monday October 27, 2025 @10:26AM
from the show-hn dept.

https://sora2video.us/video-editor/capcut

I built a simple, free online video editor that runs entirely in your browser using FFmpeg compiled to WebAssembly.No installs, no uploads, no server load — the processing happens locally, so your videos never leave your machine.Currently supported: • Trim/cut clips • Adjust playback speed • Export locally (MP4) • Privacy-safe (client-side only)It’s surprisingly fast, and works well for: •...

( Read More... | no comments | on HN )
Re:Claudeskills.cc – Share, Discover, and Reuse Claude/OpenAI Agent Sk... on Saturday October 25, 2025 @07:41AM (#45703066)
Sorry if the current version feels rough. It’s just an early stage preview — lots of improvements are on the way!
Re:Claudeskills.cc – Share, Discover, and Reuse Claude/OpenAI Agent Sk... on Saturday October 25, 2025 @01:55AM (#45701623)
I’ve been experimenting with AI agents for automation, research, operations, and workflows. One problem I keep running into is that “skills” are scattered everywhere. Everyone is reinventing the wheel, and useful prompt patterns are often locked inside private docs, screenshots, or Discord threads.

So I built https://claudeskills.cc — a simple, open, searchable place to: • publish agent skills / task capabilities • discover how others design skills for Claude/OpenAI-based agents • reuse or remix skill instructions • learn from real examples rather than theory • enable community-driven evolution

Re:A simple free tool to download social media videos for personal use on Saturday October 11, 2025 @01:15PM (#45550856)
Hi HN — I built a small web tool that extracts downloadable video files from public social media links (YouTube, TikTok, Instagram, X, etc.). It’s free to use and designed primarily for creators who want a quick, no-login way to back up their own content or grab clips for fair-use editing.

What it does • Paste a public post URL → returns a direct video download (MP4) or a high-quality stream link. • It's free;No optional credit/subscription tiers for heavy use. • Server-side fetching to avoid CORS issues and to provide stable downloadable URLs. • Lightweight: designed to handle single-file downloads and short clips; large files handled as links rather than uploads.

Why I built it • I was tired of screen-recording or chasing platform-specific quirks when I needed my own content or short clips for editing. • Wanted something minimal and fast that creators could rely on without installing CLI tools.

Important: legal & ethical note • The tool only works with public posts and is intended for personal backups and legitimate editing (fair use / creator-owned content). Please respect platform terms of service and copyright law — don’t use it to redistribute or claim others’ copyrighted works. If you’re unsure, ask the content owner or rely on platform-provided download features.

If you’re interested in trying it out or giving feedback: https://nanobananas.site (demo). I’d appreciate any technical feedback — performance, edge cases, or suggestions for safer defaults.

Happy to answer questions about implementation (fetch strategy, link signing, rate-limiting) — I can share more details if it’s useful.

— ray sun

Show HN: I tried to remove dynamic watermarks from Sora2 videos using AI

Posted by bingbing123 on Saturday October 04, 2025 @11:18PM
from the show-hn dept.

https://sora2video.us/video-watermark-remover

Hi HNI recently ran into a common issue with AI-generated videos — dynamic watermarks that move across frames. Instead of manually editing them out, I decided to experiment with AI-based cleanup and wrote a small tool called *Sora2 Video Cleaner*.It uses frame-level detection and temporal consistency to automatically identify and smooth out watermark regions, restoring a cleaner video output....

( Read More... | no comments | on HN )
Re:Sticker generator using Sticker AI and nanobanana on Tuesday September 02, 2025 @02:36AM (#45099786)
I hacked together a small project that mixes Sticker AI with nanobanana (a fast and cheap image generation model). The goal was to make it easy for anyone to generate their own sticker packs without paying a lot or waiting forever.

Right now you can: • Generate stickers in multiple styles + expressions • Export them for use in messaging apps • Run everything at low cost thanks to nanobanana

It’s still an early version, but it works — I’d love feedback on what features you think would make this genuinely useful for creators.

Re:Turning Images into Cinematic Clips –NanoBanana+Veo3 = Surprisingly... on Sunday August 31, 2025 @09:51AM (#45083151)
I tested a simple workflow: • Generate an image with NanoBanana. • Feed it into Veo3 as the first frame.

The outcome is impressive: • Very high consistency between still and video. • Smooth cinematic motion, not the usual jitter. • Feels like the image just… starts moving.

Fast. Cheap. Surprisingly good. Curious if others are exploring image → video chains like this?

Older »