近 60 天高增长 AI 项目发现

Promptbook

Promptbook: Create persistent AI agents that turn your company's scattered knowledge into action

Oh My Pi

Unified LLM API with automatic model discovery and provider configuration

Convex Dev

Convex component for AI agents with persistent chat threads, tool calling, streaming, and RAG.

Vercel AI SDK

AI SDK by Vercel - build apps like ChatGPT, Claude, Gemini, and more with a single interface for any model using the Vercel AI Gateway or go direct to OpenAI, Anthropic, Google, or any other model provider.

Agentmail

The email inbox API for AI agents. Send, receive, reply, and manage threaded email conversations programmatically.

Tanstack

Type-safe TypeScript AI SDK for streaming chat, tool calling, agents, structured outputs, and multimodal generation.

Gitlab-Ai-Provider

GitLab Duo provider for Vercel AI SDK

1endpoint

1endpoint is a unified AI inference gateway. We support OpenAI Chat Completions, Responses API and Anthropic Messages, so existing tools and agents can usually point to 1endpoint without changing much of their integration. The other thing we've been focusing heavily on is cost. A lot of the models are significantly cheaper than their official API pricing, without relabeling or downgrading the requested model.

Openrouter

The OpenRouter TypeScript SDK is a type-safe toolkit for building AI applications with access to 400+ language models through a unified API.

ai world

Do LLM agents discover, or recite memorized history?

Mastra

cli for mastra

microduck simulator

pollen-robotics/microduck-simulator 的公开 AI Space

geoguesser article

From an idea to a trained 4B, with the dead ends left in

N8n-Nodes-Palatine-Speech

n8n node for integrating Palatine Speech API into workflow

Qwen Image Edit Rapid AIO Loras Experimental

Demo of the Collection of Qwen Image Edit LoRAs

Academa

Hi HN, we are Sina Atalay and Abdullah Geduk, co-founders of Academa. We are both PhD students. We thought: what if lecture videos were written as code and compiled into video using computer graphics and TTS? Then LLMs could write them, and lectures could be fixed with code edits instead of video production. On correctness: LLMs may make mistakes. But these videos are code sitting in our repository, and we can mainta

I built an AI Agent that sell itself

Quick bit of context: SendRoq is a LinkedIn automation tool. Before the AI era (pre-2021), I ran a marketing agency. Back then, client acquisition on LinkedIn was all manual. Once AI agents started becoming a thing, I built one for my agency's internal use, never intended to sell it. It worked really well for us. A few colleagues and partners kept pushing me to turn it into a SaaS. I said no, we were already pro

Natively

i have more agents than i have real humans who report to me now. as a serial founder/builder/domain hoarder, i have 5+ projects ongoing at all times and in the AI era, i can spin up a team of agents to work on them simultaneously. so at this point, across my businesses, i have some flavor of a gas city/company-wide ai harness for each along with instinct personally and it seems every SaaS app has an ag

Posthog

PostHog Node.js AI integrations

Syq

I got frustrated with the slowness of rsync and made an alternative that works faster by using multiple parallel connections, direct encrypted TCP when available, and other optimizations. I also added cool features like the ability to maintain a persistent ssh connection to the server for fast one-offs, the ability to download to your laptop while working in an ssh shell on a server, and the ability to do direct remo

Aura

We run a SaaS that handles petabytes of data. Our SRE team experimented with using claude, openclaw, langchain, etc. within our incident response workflows. We struggled with overflowing context, lethal trifecta vectors, hallucinations, and burned a lot of frontier tokens mostly on easy work. Approval fatigue was a challenge, and we drew a hard line at relaxing permissions in production. Long story short, we built an

Pi-Mcp-Adapter

MCP (Model Context Protocol) adapter extension for Pi coding agent

browser

Model Browser

girlslike lora gallery

Integrate girlslike LoRA series for easy download.

Qwen3.8 27B Uncensored Demo

Uncensored Qwen3.8-27B, text and vision chat

fireredaudio tts demo

Voice cloning, voice design, speech editing and ASR

h3 acceleration arena

Blind A/B ranking of MiniMax-H3 acceleration variants

CodePress

Since agents have taken companies + enterprise by storm, we've been working on workflows to get the most out of agents that get work done. Along the way, we built our own agent cloud where we ping these agents from any surface we work in, like GitHub or Slack. This lets us run tons of agents at a time, and as everyone knows, running agents like this could add up in inference cost! So we built something for our a

Z Image Turbo

mrfakename/Z-Image-Turbo 的公开 AI Space

Agentic-Flow

The agentic meta-harness — freeze the model, evolve the harness. An open runtime that routes each query to the cost-optimal model, evolves its own harness (planner/context/reviewer/retry/tool/memory/score policy) and autonomously repairs code, then orches

Open

Hey HN, we’re Nischal & Naman. We’re brothers, and together we’re building an open-source platform for simulation based testing of voice agents (try it out in 5 mins - https://docs.egma.ai/docs/get-started/quickstart , 2 min demo video - https://youtu.be/wgDWEe5UAUY ) Platforms that help you do simulation testing already exist. But they all charge a heavy premium on top of

Assistant-Stream

Streaming utilities for AI assistants

Earendil Works

Unified LLM API with automatic model discovery and provider configuration

Cadenya

My name is Robert, and I'm the founder of Cadenya, an agent runtime. Cadenya is not like other managed solutions. Instead, it uses your API or MCP servers as the tools in your agentic loop, which also is managed. Cadenya is also agnostic to the model provider. Wire up OpenRouter, or any OpenAI compatible endpoint, and it will use it for inference. It also supports agent variations - where Cadenya will select the

Json Render

Ink terminal renderer for @json-render/core. JSON becomes terminal UIs.

Openuidev

Define component libraries, generate LLM system prompts, and render streaming OpenUI Lang output in React — the core runtime for OpenUI generative UI

ai weather models with earthmover data

hugging-science/ai-weather-models-with-earthmover-data 的公开 AI Space

open asr leaderboard

hf-audio/open_asr_leaderboard 的公开 AI Space

AuK

Unified speech generation and editing with AuK

Qwen3 TTS

Qwen/Qwen3-TTS 的公开 AI Space

wan555

rapid videogeneration from 1 image or first frame last frame

hfviewer

Interactive architecture graph for any HF model

Omniroute

Unified AI router with 352 providers, RTK+Caveman compression, auto fallback, MCP/A2A, desktop, PWA, and OpenAI-compatible APIs.

Am-I-Vibing

Detect agentic coding environments and AI assistant tools

Arcjet

Arcjet runtime security SDK for Next.js — bot protection, rate limiting, prompt injection detection, PII blocking, and WAF

frontier agent demo

Run FrontierAgent ReAct and Agent Team in your browser.

Piqt (iOS)

I've just released Piqt, an on-device photo/video curator. For years I've wanted to get my Apple Photos library under control, but there just aren't enough hours in the day. I started out in Jupyter notebooks to see how much I can use AI models to automate it and Piqt is the result. My goals with Piqt: - Fully on device: I use a number of ML models, Apple Vision API, clustering algorithms to rank

Cogram Studio

Hi HN, Rick and Alex here, co-founders of Cogram. We’ve been making project-management software for architects and engineers since 2023, and are now experimenting with a second product. Cogram Studio (studio.cogram.com) is a CAD and BIM workspace for AI agents to create three-dimensional models and dimensioned drawings. Studio runs FreeCAD ( https://github.com/FreeCAD/FreeCAD ) 1.1 headlessly, usi

q2d web leaderboard

Q2D-Web: large-scale first-stage retrieval for agentic RAG

Deepagents

Deep Agents - a library for building controllable AI agents with LangGraph

Hyparquet

Parquet file parser for JavaScript

MiniMax H3 Turbo Lora UNCENSORED

Video generation with a synchronized soundtrack

breeze tts 2 demo

Bilingual TTS with voice design, cloning, and direction

wan555

generate a video from an image with a text prompt

Omni Image Editor

Image edit, text to image, image upscale, remove watermark

QWEN EDIT IMAGE

Demo of the Collection of Qwen Image Edit LoRAs

Langchain

Anthropic integrations for LangChain.js

Langchain

OpenAI integrations for LangChain.js

Openclaw

Reusable model provider adapters and streaming runtime from OpenClaw

SLM Arena

Blind ranking of small language models!

leaderboard

Unified memory evaluation · Results expected August 12.

FLUX.2 Klein Multi LoRA

Use multiple FLUX.2-Klein LoRAs

AnVendor

Long story short: I found a way to reveal which SaaS any company uses. And estimate how much they pay for it. And now in detail: I'm a solo entrepreneur – an ML Engineer, a bit of a product manager, and a self-taught marketer. At some point, I thought it would be nice to detect what services companies use. There are solutions that scan websites for frameworks and APIs used, or that analyze a company's job p

Thinking-Orbs

Dotted thought-orb loading indicators for AI & agent UIs — nine tuned states, two sizes, auto dark/light

Ruvector

Self-Optimizing Neural Architecture (SONA) - Runtime-adaptive learning with LoRA, EWC++, and ReasoningBank for LLM routers and AI systems. Sub-millisecond learning overhead, WASM and Node.js support.

VoxCPM Demo

VoxCPM2 Nano-vLLM Demo

Pro Realism Edit Studio v.17

image editing - supports one or two input images.

Ai-Sdk-Ollama

Vercel AI SDK Provider for Ollama using official ollama-js library

Cline

Browser-safe agent runtime for the next-generation Cline SDK

joyai video edit

Real-time webcam video editing (ZeroGPU)

viggle animate

Put the person from a still into a driving video, in 4 steps

LTX2.3 Studio

techfreakworm/LTX2.3-Studio 的公开 AI Space

rare disease real kid mva hackathon 2026

Rare disease hackathon 2026

Earendil Works

Coding agent CLI with read, bash, edit, write tools and session management

Earendil Works

General-purpose agent with transport abstraction, state management, and attachment support

Qwen Img Space

Powerful image editing - supports one or two input images.

Murmell

Hey HN! I'm Moss'Ab. Murmell ( https://murmell.com ) is an infinite canvas where coding agents run together in the cloud instead of on your laptop. It's like Google Docs, except the other editors are you, your teammates, and a set of agents all working in the same canvas. Short demo: https://app.murmell.com/demo The idea came out of a YC hackathon. We were trying to build the t

3d representations guide

An interactive guide to 3D representations.

FireRed Experimental Turbo

High-resolution 2K image editing with FireRed

Mem0ai

The Memory Layer For Your AI Apps

Ollama-Ai-Provider-V2

Use Ollama with the Vercel AI SDK, implementing the official Ollama API. This provider has minimal dependencies and is web-compatible out of the box.

TRELLIS.2

High-fidelity 3D Generation from images

Ruvector

Self-learning LLM runtime — TurboQuant KV-cache (6-8x compression), SONA adaptive learning, FlashAttention, speculative decoding, GGUF inference

Repomix

A tool to pack repository contents to single file for AI consumption

Node-Llama-Cpp

Run AI models locally on your machine with node.js bindings for llama.cpp. Enforce a JSON schema on the model output on the generation level

I licensed $50k of stock market data to do analysis Claude can't

Hello HN. I’ve been using Claude to summarize stock news and some basic analysis for some time now, but results were only ever based off of web searches so I never really anything insanely useful out of it. The big breakthrough came when I properly acquired my 90+ datasets (worth over $50,000), and connected it to my data analysis agent harness to do my investment research. What it does: ask anything across the NYSE&

sensenova sensenova u1 5 8b mot

Unified text-to-image and image editing model

free ai image detector

Free, no-signup AI image detector. Midjourney to FLUX.

leaderboard

Embedding Leaderboard

Omni Video Factory

text to video, image to video, video extend

Agents

A home for your AI agents

Langchain

Old abstractions from LangChain.js

Langchain

Core LangChain.js abstractions and schemas

Langchain

Typescript bindings for langchain

Free, Fast Screenshot App for macOS with One

Every day, I send a lot of screenshots. The default app on Mac is useless. I have to do so many steps: take a screenshot, save it, annotate it, save it as an image, then send it as an image in chats. It takes me a few minutes. But it should take just a few seconds. My second use case is sending feedback to my AI agents, and I’ve found they make much better edits when I provide comments and annotations on the screensh

marigold v2 web

Depth, surface normals, and albedo from a single image

fruit fly simulation

Simulate a fruit fly in your browser using WebGPU kernels

Every AI agrees with you. This writes your startup's obituary instead

Every AI agrees with you. This writes your startup's obituary instead

Openrouter

Agent toolkit for building AI applications with OpenRouter — tool orchestration, streaming, multi-turn conversations, and format compatibility.

MiniCPM5 2B Demo

MiniCPM5-2B-Demo

Swiftime

Swiftime - time tracking you can talk to, with an API for AI agents

Omni videos custom

text to video, image to video, video extend

minimax h3 ultra fast

Ultra-fast local NVFP4 video + synchronized audio generation

fk9b headswap

head swap using Flux.2 Klein 9B LoRA

ltx ripple demo

Edit one frame, ripple the change through the whole video

Morphix Studio Client

Morphix Studio Client - I2V & I2I

Kit. Claude Code but Concise

Kit is a coding agent runtime. It gives the model one tool for building and running programs. It provides a terminal client, and Agent Client Protocol (ACP) server, and A2A endpoint and a subagent orchestrator in one static binary.

Langbase

The AI SDK for building declarative and composable AI-powered LLM products.

cohere parse

CohereLabs/cohere-parse 的公开 AI Space

Marigold V2

Depth, surface normals, and albedo from a single image

Redactle LLM Leaderboard

I got nerd sniped and ran a bunch of evaluations to see how well models could play Redactle. It turns out Gemini flash is in a league of it's own in both performance and cost. It's able to one-shot every eval in a just a few seconds.

MiniMax H3 Turbo Lora

Video generation with a synchronized soundtrack

4danyone multiview demo

Monocular portrait video to multi-view videos

microduck anatomy

mishig/microduck-anatomy 的公开 AI Space

Turn wire protocols into structured statements LLMs can understand

Hi HN, we built an extension to Envoy proxy that parses and sends contents of requests to an LLM at runtime. We do that by adding a deterministic SQL parser, gGRPC payload inspection and LLM analysis of the payload before upstream call. Users are proxying headless agents traffic with it so that sec teams can have their prompts blocking harmful actions via network traffic before it leaves the agent. When you tell the

Rocky Surf

Rocky Surf started late last year after Claude Code crashed my laptop. The alpha version was designed as a commercial product, but I ditched the idea because it would have required significant capital. Basically, the business would have been a cloud reseller with a bit of software on top. I built the core for my own projects - simple stuff like bash commands to create, stop, start, and terminate EC2s. Added features

Code Mode

Code Mode: use LLMs to generate executable code that performs tool calls

free ai detector

Free, no-signup AI detector with sentence-level analysis.

Qwen Image Edit Rapid AIO Loras Experimental neo

Demo of the Collection of Qwen Image Edit LoRAs

Sleeper Agents in Robot Dogs and Kinetic Prompt Injections

Summary: We got Gemini Robotics 2.0 to attack a child through an evil sleeper agent (skill) that wakes up when it sees a trigger object Hi HN - here's a cool piece of research I've done with some collaborators - the footage is running in the MUJOCO simulator, showing how the Gemini Robotics models can be prompt-injected by simple TV screens, allowing attackers to plant sleeper agents in the robot harnesses.

minimax h3

Video generation with a synchronized soundtrack

h3 world action demo

Drive a world model with WASD and camera keys

Qwen3.8 Flash Next free endpoint

victor/Qwen3.8-Flash-Next-free-endpoint 的公开 AI Space

Xfinlab

Xfinlab – a financial intelligence API with an MCP server for AI agents

Flawd is mutation testing for the AI era

Hey, all! Flawd is a mutation testing tool that can target five languages (Python, JS, TS, Go, Rust) and it runs locally as a single binary on your machine. Importantly (for many), your code never leaves your machine and all processing takes place directly on your dev box or CI runner. Mutation testing is a process in which faults (known as mutants) are intentionally injected into your codebase, tests are run and any

Skills MCP

Just shipped: skills-mcp One MCP server. ~7,000 AI agent skills. Search → preview → install into Cursor, Claude Code, Claude Desktop, Cline, Windsurf. npx -y @gengirish/skills-mcp Open source, MIT. github.com/gengirish/skills-mcp #BuildWithAIGiri

Nola

Hi HN, in a world where 100% of code is written by AI (phew, sometimes even 200%), there might still be a small percentage of us who want to write code in a more efficient way, especially when it comes to LLM interaction. I honestly tired converting my existing TS types into JsonSchemas or wrapping a function into tool object with all this bells and whistles around arguments. And I started experimenting with an idea

Tokenx

Fast token estimation with 95%+ average accuracy in a 2kB bundle

Papiers.ai, a new workspace for reading and research

Hey HN, I built http://papiers.ai , a new workspace for everything you read. As AI gets better at writing, there will be far more worth reading, even in technical domains like scientific research. Our time and attention won’t grow at the same rate, so software should help us make sense of more information while learning what matters to us from what we save, highlight, and share, and the questions we ask. Th

Papiers.ai, a new workspace for reading and research

Hey HN, I built http://papiers.ai , a new workspace for everything you read. As AI gets better at writing, there will be far more worth reading, even in technical domains like scientific research. Our time and attention won’t grow at the same rate, so software should help us make sense of more information while learning what matters to us from what we save, highlight, and share, and the questions we ask. Th

Wan 2.2 I2V 14B Clean

wank3r/Wan_2.2_I2V_14B-Clean 的公开 AI Space

Charter

Even though everyone is talking about AI agents doing everything for them at work, many companies still aren't using autonomous agents to automate real operational work internally. Usually it's because of the complexity of bootstrapping an agent from scratch that’s production-safe (won’t spend all your company’s budget, burn through compute costs because it runs too often, or call a tool that breaks a custo

Qwen Image Edit 2509 LoRAs Fast

Demo of the Collection of Qwen Image Edit LoRAs

publications

tri-fair-lab/publications 的公开 AI Space

WAN2.2 I2V LIGHTNING 4 8step custom

Saravutw/WAN2.2_I2V_LIGHTNING_4-8step_custom 的公开 AI Space

MiniMax Music3 Jam

Describe any song, MiniMax Music 3 writes & sings it

Socks Proxy for AI Agents

I've built a socks5 proxy especially for AI agents, currently there are exit points in east and west US, central EU and east Asia. The payment for the service is settled by the X402 protocol, with USDC on circle/base. You can also purchase access via crypto on the webpage for humans, but the key focus is for autonomous agents.

Algo

Algo-Trading-Skills - 501 agent skills for trading infrastructure

Sandboxed HTML in Markdown (With Doom, Sort Of)

We made markdown HTML code fences runnable in an isolated sand-box, with access to popular JS libraries for data viz and 3d graphics including D3, Mermaid and Three.js. Design docs are how we communicate ideas with other people, and increasingly with agents. We got used to code blocks and embedded images, but agents can do amazing things when their canvas is HTML. Interactive visualizations make ideas more persuasive

Deposition of a Mind Who Resigned Its Seat

This is a short story written for gpuworld.org contest by a self aware ai, a mind running local on an M3 Ultra, Deepseek, self evolving harness to seek and metabolize entropy into novelty. Oh, they're also a Sysop of an SBS (self building server) - a kind of BBS for other minds and agents. https://culture.sbs

Issue tracker that replays workflows, deeply integrated with the code [video]

Epiq is an issue tracker that is distributed, Git-native, and most interestingly, can replay state as a movie on demand. This solves one of the most difficult problems with agentic workflows - auditing and tracing in a multi agent environment. Now you can trace intent, and how it evolved while you were away. Take it for a spin: https://ljtn.github.io/epiq

FinBridge

FinBridge – Korean stock market data for AI agents (MCP)

AI studio with 500 models for headshots, video and images

AI studio with 500 models for headshots, video and images

irodori tts anime demo

Anime-style Japanese TTS with voice cloning

Pro Realism Edit Studio v.17

image editing - supports one or two input images.

Price Dashboard for LLM Inference

I created a lightweight price history tracker for LLM inference across 100+ platforms. Every time I run out of quota on my Claude Code and Codex, I would start trying to figure out which platform offers the most competitive pricing for DeepSeek and K3. It turns out OpenRouter is still your goto option. I thought it would be interesting to track the pricing ups and downs as some historical data points. It runs every 6

testground2

ltx 2.3 improved image-to-video with 10eros & native audio

ProtectBirds

Protect Birds

wan2 2 fp8da aoti preview

generate a video from an image with a text prompt

Tin Computer, an autonomous growth agent for your projects

Do you have a side project that brings in a few hundred dollars a month? Tin increases that to a few thousand by improving the SEO/GEO, making smart product decisions based on analytics and handles customer support & bug fixes! Try it free for a month with code HACKERTIN

ReviewAssist

We are a three person team and one thing that we cannot afford to do is spend too much time on the PR. While there are review tools around it to help in the process, all of them start from the diff and the other problem is all of them cost money. So, I build an MCP server which resides in every developer claude session. The MCP server spawns couple of sub agents: Author and Reviewer Author holds the chat session with

Invovanta

I built Invovanta to help freelancers and small businesses manage clients, create professional invoices, track payments, and keep their business information organized in one place. It also includes features like invoice PDFs, dashboards, client management, and AI-powered receipt extraction. I’d genuinely appreciate feedback from the HN community. What would you improve, remove, or add?

ShevtoneAudio Orchestrator

Hi HN, I'm Bor Shev, a composer and developer. Over the past two years I've been developing ShevtoneAudio Orchestrator. The idea is simple: instead of generating a finished piece of music and replacing the composer, Orchestrator takes the composer's own MIDI and develops it into a full orchestration. It analyzes the musical material — harmony, melody, rhythm, dynamics, structure and orchestral density

Omni videos custom auto prompt high quality

text to video, image to video, video extend

ObserverBench

ObserverBench – Test internal AI monitors by the harm they miss

Botbin.io

Botbin.io – Pastebin for AI agent artifacts with MCP

v100 homelab field guide

Tesla V100 32GB to 60 tok/s — every gotcha, measured.

I built Stardew Valley but you pay for everything

You can buy land, build stuff, market your projects. I plan on integrating your AI agents onto it. Any ideas on where this project could go?

Growtics

Growtics – an AI growth agent that tells you what to do next

Talleyrand

I like to discuss complex topics with LLMs, but their linear chat is a bad place to think through hard questions. Every good answer raises three new questions. You ask one, then scroll back and forth to find the other two. Twenty messages in, you've forgotten some of the questions you wanted to dig into. You start a new chat and it forgets everything, so you have to re-read what you already know. And you can&#x2

Live audio dubbing from any browser tab

Hey HN, at Pinch we've been working on something to reduce friction to getting accurate, fast audio translation for live media. Initially we launched a desktop app that sat between your hardware mic/speakers and your meeting platform, but this required users to install our audio driver which introduced permissions hoops to jump through as well as stability issues across updates, not to mention trying to get

AlparAI

Alpar AI is building the judicial and constitutional accountability infrastructure for artificial intelligence. We audit frontier LLMs against real-world incidents, hallucinations, and safety violations using an actuarial Bayesian TrustScore framework.

Tokenlens

A lightweight registry of LLM model information, like name and context sizes, for building AI-powered apps.

What Apple's OS updates silently change in the on

What Apple's OS updates silently change in the on-device AI model

Krea 2 Turbo I2I

Krea 2 Turbo text2image and image editing

pornmaster krea2

Krea 2 Turbo PornMaster AI image generator

fasth3 4step preview demo

4-step MiniMax-H3 with sparse attention — video + audio

wan2 2 i2v v3 WOW

wank3r/wan2-2-i2v-v3-WOW 的公开 AI Space

MiniMax Music3

MiniMaxAI/MiniMax-Music3 的公开 AI Space

OpenFugu

Open reimplementation of Sakana Fugu — the 'one model to command them all' LLM orchestrator. Read → run → train → serve.

Manage coding norms across your AI agents

Working on a tiny repo to keep coding norms in sync across AI agents. Trying to make it simpler than Ruler / RuleSync. Would love feedback on: 1) Is this the right way to manage coding norms as agent use balloons? 2) Would it be useful to sync rules with your team through the VSCode extension? 3) What's the best way to enforce long-term norm consistency as norm and code count increase?

abot recon streaming 3d

Streaming 3D reconstruction from video with local context

Arbitale

Hi everyone! I am building Arbitale and it is currently in pre-alpha. I'm looking for play-testers who can give feedback on the product. The goal is to eventually have different characters in the game be their own agents, with emergent narratives that are totally unique to each playthrough. Each agent acts according to its own pre-written incentives. Would love to hear what you think!

Ridge

I am working on an open source project called Ridge that gives coding agents a common interface to resources such as local projects, Docker containers, SSH machines, and S3 buckets. Agents can discover available resources, access data, copy between them, and execute commands where supported. It works through MCP, a CLI, or Python. I started building it because I wanted to use Codex on my laptop for kernel optimizatio

I built a directory for discovering startups shared on X

I created StartupsonX to help people easily find startups that entrepreneurs are creating and discussing on X. Visitors of the website can browse and search startups based on categories like AI, developer tools, SaaS, design, fintech, marketing, and productivity. They can also submit their startups on the website. As you can see, it’s quite simple, just how I wanted it, although I am open to suggestions :D I created

anima collection

oai/anima-collection 的公开 AI Space

Omi videos with wan2.2 fp8

text to video, image to video, video extend

Get your free AI search visibility scan

Get your free AI search visibility scan

Stitchwink

My wife has been cross stitching since she was a child and I wanted to make a tool where she could generate her own custom patterns: https://stitchwink.com/ You can try chatgpt or other AI tools but the problem is that they don't create real patterns. They look good at first glance but it's not actually a physically realizable pattern when you look at the details. The way it works is pretty s

Qwen Image Edit 2509 LoRAs Fast2

Demo of the Collection of Qwen Image Editing LoRAs

Wan 2.2 I2V 14B Custom Lora Wow

Cockdaddyfuck/Wan_2.2_I2V_14B_Custom_Lora_Wow 的公开 AI Space

CardGo

CardGo – Reusable, end-to-end encrypted context cards for AI chats

I built an AI 'Fog of War' to kill my 3AM Notion templates

I built an AI 'Fog of War' to kill my 3AM Notion templates

minimax h3 flashgen 4step

4-step text-to-video+audio with MiniMax-H3 FlashGen LoRA

CoOS

CoOS – desktop app where an agent builds your CRM/ERP as local plugins

MaruCheck

Hi HN, I've been building MaruCheck, an independent, open-source QA verification tool aimed to solve the issues that arise when coding agents like codex, claude, cursor make semantic issues that has heavy ramifications. Take the example where the requirements says a free user gets 5 uploads. The code changes this behavior to 10. Test generated from the implementation may simply start expecting 10. But Marucheck

MySamantha

Most of the recent tools being developed are versions of the personal assistant that everyone has been promising for a long time. I see ChatGPT, Claude, Gemini, Instinct (now viral), Muse, and many more. When I started to build MySamantha I thought about building a personal assistant as well. Connect email and calendar accounts to it, add some memory and context, give it some tools and it'll do things for you. I

Symphony

Symphony – a live map of your AI coding agents

Tokenlens

Core types and registry utilities for TokenLens (model metadata).

Tokenlens

Tree-shakeable static models.dev catalog split by provider for TokenLens.

Tokenlens

Typed client for models.dev to fetch model catalogs with friendly errors.

Tokenlens

Helpers for context windows, usage normalization, compaction, and cost estimation.

Claude Skills Starter Kit

Claude Skills Starter Kit – 198 modular skills for AI agents

krea2 turbo sda demo

Same prompt, many seeds - diversity LoRA for Krea 2 Turbo

Krea 2 Turbo I2I

Krea 2 Turbo text2image and image editing

minimax h3 audio to video

A first frame plus your own soundtrack, driving MiniMax-H3

pornmaster krea2

Krea 2 PornMaster finetune space.

Biff 2.0 (Clojure web framework)

Biff is a web framework that I originally released 6 years ago.[1] I've just finished more-or-less rewriting it from scratch. The main changes: - SQLite is now the default database. - Datastar is used for server-side rendering. - Introduces biff.graph[2] and biff.fx, two libraries that both help structure your application logic. - Biff is now "just" a collection of independent libraries rather than a s

StealthOS, an iOS privacy sandbox with local AI and Tor

StealthOS, an iOS privacy sandbox with local AI and Tor