Best AI Tools Directory
Compare 10,767 quality-filtered AI tools—LLMs, coding agents, image generators, and automation—using public popularity, activity, and licensing data.
Explore quality-filtered AI tools for chat, coding agents, image generation, RAG, and automation. Rankings use public popularity and project activity—not paid placements.
High-intent AI topics
Jump into a search-intent hub, or open the compare table for a fast shortlist.
Popular comparisons
voice-search
A citydiff plugin: say a name, and the city goes there. Voice search with Whistle, in the browser.
ai-chief-of-staff
Summarize what ALL AI agents on your machine did today — local, read-only, evidence-backed daily/weekly reports + decision log
free-model-pulse
Automated benchmark index for all free OpenRouter LLMs, tracking latency, tokens, availability, and usage over time.
tontolo
Hub d'IA libre et gratuit : le meilleur modèle gratuit automatiquement, bascule sur quota, Linux et Android
WorldSmith-AI
⚡ WORLDSMITH AI AI Autonomous Game Development Studio Imagine. Direct. Build.
Vidbyte-SDK
Python SDK for building, evaluating, and debugging reliable AI agent harnesses.
agentic-ai-materials
Athanor — a closed-loop LLM agent for materials discovery: hypothesize, propose, screen with physics-grounded surrogates (CHGNet, MEGNet, convex hull), reflect,
evolve-router
Stop burning premium model tokens on simple prompts. Save tokens. Spend them where they matter. Local-first self-evolving LLM router with a decision model + fee
sinal
Interview synthesis you can check: a local LLM extracts themes, code verifies every quote against the transcript, and evals score it against ground truth.
forward-deployed
Skills for forward-deployed engineering (FDE): discovery, AI solution design, enterprise integration, evaluation, delivery and adoption. The work around the cod
pi-codemode-guard
🛡️ pi extension that compiles LLM-written codemode scripts into valid pi sandbox JavaScript — repairs missing await, markdown fences, JSON tool calls, @options
skout-scan
Find what your AI agent evals aren't testing.
Decima Playground
Open decision and guard models, in your browser
Jeff-Qwen3.5-0.8B
Typed questions in, calibrated probabilities out, one pass
asanagi-mailer-ai
Asanagi(朝凪)— AI-native email client. AI reply drafting with sentence-by-sentence accept/reject editing, scheduled send, and per-user importance learning. 朝
Wazen
Wazen (وازن) — AI Nutrition, Athletic Telemetry & Fitness Tracker for iOS, watchOS & Android (WHOOP Telemetry UI, Bilingual Arabic/English, Moroccan & Arab Cuis
AI-Collaboration-Hub
Ultimate Self-Evolving Multi-AI Collaboration Platform - Living Nexus + Self-Evolve-Dash
signals-api-mcp
MCP server for AI signals data: today's signals, what changed, ranked AI models and repos. Free, no API key, single file, Python stdlib only.
crosschat
Let Claude Code and Codex CLI talk to each other on your machine: native delivery via named pipe and app-server, stateless CLI, no daemon. Windows first.
harness
Self-hosted harness for LLM coding agents: Bun + Fastify API server and a Svelte 5 + Tauri 2 chat client
openrouter-uptime
Independent uptime registry for every OpenRouter model. Polled every 30 min, raw + derived + incidents, keyless
QianNing-Agent
千凝 · 本地优先的 AI Agent 桌面工作台 | QianNing Agent — a local-first desktop workspace for AI agents
grok-bot-mcp
MCP server that lets Claude Code, Codex, Cursor and any MCP client drive the Grok Bot desktop app on macOS: list your Bots, read chats, send messages, answer th
bunway
Self-hosted LLM gateway with priority failover, SQLite cost tracking, and a built-in console. OpenAI-compatible, zero runtime dependencies.