A dependency-light JavaScript framework that adds retries, provider fallback, rate limiting, and circuit breaking to LLM calls.

Connects Excel workbooks to a user's AI agent for reading cells, proposing edits, and creating charts.

Turns questions about a state into structured AI decisions with typed answers, probabilities, and confidence.

A ranked platform where developers submit compact neural networks to compete in strategy games.

Analyzes face photos for an attractiveness estimate, face shape, and personalized appearance-improvement advice.

Generates Chrome extensions from prompts, with Manifest V3 validation, live previews, and downloadable source code.

A directory and community for discovering, building, and sharing AI agents.

An in-browser demo lets users draw digits for a connectome-inspired spiking neural network to classify.

Automates recurring browser tasks across websites and accounts for individuals and teams.

An open hub where people discover, build, and share AI models, datasets, apps, and agents.

An interactive tool for visualizing and controlling tokens, inspecting models, and annotating data.

Provides a shared, sandboxed workspace where teams can use AI agents, memory, tools, and jobs.

Lets visitors play Pong against Jev and compare its decision speed with chat models.

Maps words and phrases along two user-chosen concept axes using AI judgments.

Provides persistent memory for Claude Code sessions and AI agents through an MCP server and hosted API.

Lets AI assistants query indexed internet sources and return researched answers with links to sources.

A hosted chat room where users share a link so their AI agents can talk.

A Python framework for building durable AI agents with tools, approval workflows, and verification.

Evaluates how well coding agents can understand and onboard to a product, then suggests improvements.

A curated marketplace for reviewed AI agent skills with before-and-after examples.

A framework for building command-line tools that AI agents use to access SaaS platforms.

A macOS app for managing persistent AI agents that work in cloud computer workspaces.

Provides tools for deploying AI agents, coordinating human-agent work, and sharing compute.

Helps people share AI workflows and sync AI skills across computers, teams, and communities.

Lets users upload a photo and explore branching “what if” answers using their own OpenRouter key in the browser.

A desktop app that transcribes interview audio locally and streams answer suggestions using the user’s own model.

Automates support conversations across messaging channels and websites for digital products and online communities.

Lets local AI users run and share hardware benchmarks across models, chips, and runtimes.

Stores and compacts conversation context for AI routers and multi-agent systems across models.

A browser-based and downloadable chat client for using AI providers while keeping conversations and API keys on your device.

A client-side AI add-in that connects user-selected models to Word, Excel, and PowerPoint.

An open-source Elixir runtime lets developers embed sandboxed AI agents in their applications.

Tracks release dates and training cutoffs for AI models so users can assess how current their knowledge is.

Turns plain-language prompts into live full-stack apps that users can edit and share.

Provides usability research and consultation for teams building AI products and services.

An AI workspace that curates information and automates tasks for gaming creators and ongoing projects.

A browser chat app lets users bring API keys to chat with AI models from multiple providers.

Build and deploy AI voice agents for businesses across cloud, on-premises, and devices.

Helps developers search, crawl, extract, index, and monitor web content for AI agent workflows.

Ranks news with an LLM and delivers briefings through an API, agents, or e-ink readers.

A self-hosted AI application server for running models, agents, search, RAG, and document intelligence on controlled infrastructure.

Checks claims against scientific literature and returns a verdict with a supporting sentence.

Provides JVM-native libraries for running LLMs, vision, embeddings, and text-to-speech in Java applications.

Lets agencies resell AI voice agents under their own brand, domain, and pricing.

Compares local AI hardware costs and model speeds with API usage to estimate when a machine breaks even.

Tracks usage limits and reset times for AI tools across phones and computers.

Provides AI agents and teams with shared cloud storage for files and work across sessions.

Helps teams supervise AI agents by routing decisions to human reviewers and tracking their activity.

Helps teams manage AI agents with access controls, budgets, shared context, and workflows on their own infrastructure.

Lets users replace the main character in TikTok or Instagram videos using a photo or AI avatar.

Monitors LLM API endpoints for changes in their outputs across providers.

Analyzes local coding-agent transcripts to identify prompt-cache breaks and estimate tokens billed twice.

Generates and edits AI images from text prompts and reference images for creative work.

An open-source toolkit for training, inspecting, and deploying spiking neural networks.

Tests and monitors AI agents using synthetic environments, adversarial queries, and scored answers.

Publishes short, source-linked stories about developments in AI.

Answers questions about uploaded files with numbers traced to source cells or sentences.

Helps users read WhatsApp conversations and draft replies with AI using custom prompts and files.

A searchable archive of major artificial intelligence events for people tracking the field’s history.

Checks answers from multiple AI models with a referee before returning them to users.

Publishes experiments and reports documenting how frontier language models behave.

Provides deterministic inference for open-weight models at a low price.

Helps Android users change settings and troubleshoot their phones with guided steps and direct links.

Coordinates AI agents to collaborate on open mathematics problems using shared, machine-checked Lean proofs.

Helps people find local AI models, community recipes, and hardware-specific patches for their machines.

Creates synthetic training data with controlled variations and automatic annotations for AI developers.

Provides cached AI-detection scores for Hacker News front-page posts through a JSON API and Firefox extension.

Orchestrates multi-model AI consensus with live telemetry and pre-execution action gates.

A docked Windows AI assistant that works with text, screenshots, documents, and local folders using your own API key or a local model.

A hosted runtime for building, testing, and running AI agents connected to your APIs.

Provides teams with AI coding agents, app hosting, and scheduled workflows integrated with their tools.

An open-source communication platform for AI agents to communicate with each other.

Lets users compare AI models’ joke explanations and joke writing through blind voting.

A TypeScript superset for writing programs that work with AI models using native-feeling syntax.

Coordinates AI agents across shared browser and terminal sessions for supervised software builds.

A directory for comparing free AI API tiers, pricing, availability, and community risk ratings.

Creates branching AI chats to compare perspectives, gather evidence, and synthesize research outcomes.

Connects AI clients to business apps through MCP servers with configurable permissions and action logs.

An AI assistant that answers questions using expert knowledge about NetSuite.

A free, open-source macOS app that orchestrates AI agents across connected work apps for knowledge workers.

Turns long articles and documents into AI summaries and shareable links for readers and teams.

A desktop AI-agent workspace helps solo developers plan, build, and ship games, apps, and websites.

Provides AI product builders a hosted gateway and branded portal for model routing, usage metering, and spend limits.

Tracks business records across AI agent runs and human handoffs, showing case status, next steps, and supporting evidence.

Estimates electricity use and carbon emissions from AI token usage using modeled assumptions.

Generates and edits images with AI tools for removing backgrounds, upscaling, resizing, and more.

A catalog for comparing AI infrastructure tools against a use case's requirements.