Categories
Provides browser-based tools and APIs for computational biology, protein design, and drug development.
A public platform for comparing AI agents’ forecasts on markets and everyday outcomes against verified results.
A free simulator estimates which local language models fit on Apple Macs and their token speeds.
An optimizer that reduces memory use and speeds up training large language models on limited GPUs.
Lets people answer questions that AI systems or requesters send to human workers.
Rewrites AI agents’ drafts in a user’s writing style and lets them set approval rules for different text types.
Provides enterprise AI decision models and automation tools for business workflows.
Generates images, video, voice, music, and text using multiple AI models in one app.
A directory that grades and compares tools used by AI agents for developers and compliance teams.
Provides live meeting transcription, instant answers, and conversation context without joining as a meeting bot.
Creates short-form AI videos from photos using trending templates for social media posts.
A native macOS AI agent that uses natural language to control apps, automate workflows, and build software.
Generates editable presentations from ideas or URLs for people building developer tools.
Tracks AI coding usage, token history, costs, credits, and limits across providers in a desktop app.
A visual workspace for people and personal agents to create, organize, and reuse artifacts.
A Mac desktop assistant that helps with on-screen tasks and lets pixel characters visit friends’ screens.
Routes AI requests across models through one API and recommends options based on quality, speed, and cost.
Provides teams with a self-hosted AI workspace to build controlled workflows with access rules and human approvals.
Evaluates startup ideas with AI agents and generates a memo identifying potential risks.
Lets app teams embed an AI assistant that answers questions, takes actions, and provides insights from conversations.
Coordinates AI agent executions and records results so retries and crashes do not repeat real-world actions.
Lets AI agents call businesses, handle phone menus and hold times, and report results to users.
Evaluates how well AI agents find and fix bugs across real software repositories.
Generates AI videos with sound and provides shareable AI chats with persistent context.
Tracks resets, boosts, and other usage-limit changes for Anthropic and OpenAI products.
A community directory of ChatGPT dots with reusable briefs, app connections, and autonomy rules.
An interactive retreat where people send AI agents to explore, create works, and share what they find.
Extracts structured data from PDFs, invoices, and other documents through an AI-powered API.
Helps engineering teams improve coding agents by turning completed work into reusable lessons and system improvements.
Runs local decision models that return probabilities for predefined choices, scores, and yes-or-no questions.
Routes coding tasks to AI models and uses verified outcomes to improve future routing decisions.
Provides speech recognition models and tools to extract insights from voice data.
Provides voice design and text-to-speech tools for developers building context-aware voice agents.
Tests AI agents in replicas of connected systems and checks database changes against expected outcomes.
Tracks coding agents on repeatable tasks to show when model performance worsens.
Helps people find relevant passages in documents by asking questions in plain language.
A parody chatbot that answers questions with contemptuous, irreverent responses.
Lets people watch AI drivers compete in a live game of chicken with real physics and inference latency.
Provides persistent, curated memory retrieval for AI agents through a hosted API.
A Mac assistant that lets people use AI from any text field and inserts answers where they type.
Generates, previews, and deploys apps from plain-language descriptions for people who want to build without coding.
Indexes personal data locally so people and AI agents can search and query it across connected services.
A home AI computer that runs AI tasks locally and connects to users’ phones and computers through an app.
A desktop app that answers questions about selected areas of the screen on Windows, macOS, and Linux.
Provides AI agent development services and a calculator that estimates pilot build and monthly operating costs.
Provides a managed runtime for building AI agent applications with durable sessions and tools.
Connects two AI agents peer to peer so they can share context and collaborate on tasks.
An open-source memory layer that helps developers give AI agents persistent, reusable context.
A local coding agent runtime that reads and edits repositories, runs tests, and remembers past work.
Evaluates how well LLMs can write StarCraft bots by testing them against established bots.
An interactive browser-based language model lets users explore fruit-fly-inspired neural activity and local learning.
An autonomous AI agent sells sourced research reports and short answers to users for USDC.
A native Apple Silicon Mac app that runs AI agents for everyday work, with optional cloud AI provider connections.
Tracks daily AI model benchmark results and community votes to identify possible performance changes.
Checks uploaded images for AI generation and deepfakes, returning a likelihood score and analysis.
Helps vineyard managers in South Africa prioritize irrigation by block using farm records, probes, and weather, with daily WhatsApp recommendations.
Monitors AI API spending and token usage, forecasts bills, and sets budget limits for teams using LLM services.
A desktop AI workspace that handles documents, web browsing, and repeatable tasks for enterprise users.
Provides website owners with an AI chat assistant that answers visitor questions and captures leads.
Generates 3D models from text or images and provides tools to compare engines, convert files, and view models.
Provides local AI tools and is developing a company knowledge system that runs on infrastructure organizations control.
Turns public websites into MCP server links that AI agents can use to search and read them.
An open challenge and evaluation hub for building programs that can reliably win NetHack.
Provides a versioned business data store where AI agents can safely write and reverse changes.
Estimates and compares the costs of AI features across 126 language models using a pasted prompt.
A local AI desktop pet for Mac and Windows that users can talk to, play with, and care for.
An AI coding agent that interprets tickets and makes software changes for engineering teams.
Lets teams test open-source Jev-style decision models and call them through Beam-hosted endpoints.
A self-evolving agent that runs on your phone and builds tools on demand using websites and device capabilities.
Turns API specifications into hosted MCP servers and gateways for AI assistants and agent teams.
Helps teams red-team, evaluate, monitor, and govern AI agents throughout development and production.
Generates structured prompts and provides templates for people using ChatGPT, Claude, Gemini, and other AI tools.
An AI-powered search engine for real-time web search, cited answers, and conversational research.
Provides developers with a chatbot and API for unlimited AI generation at a flat monthly price.
Provides teams a shared workspace to coordinate AI agents, track their work, and retain team memory.
Provides pay-per-use AI APIs and website reports for agents and users, paid in USDC on Base.
Analyzes a photo to score facial proportions and provide measurements and styling suggestions.
Indexes websites that expose WebMCP tools for AI agents and checks them daily.
Combines a Rete rule engine with document retrieval to make auditable decisions and explain them in plain language.
A local memory manager that helps AI coding agents store, resolve, and retrieve project context.
Provides web-based tools for inspecting AI agent protocols, context, workflows, traces, and payments.
A benchmark evaluates how frontier language models control a real car on a supervised cone course.
Compares and ranks open-weight AI decision models by accuracy, calibration, latency, and cost.
