All AI tools
1,521 tools across 18 categories.
Open-weight reasoning models, free to chat with
DeepSeek's R1 and V3 models trade blows with frontier labs on maths and code at a fraction of the price, and the chat
The most natural AI voices
State-of-the-art text-to-speech and voice cloning with emotional range, used for audiobooks, dubbing and agents.
The translation quality benchmark
DeepL still produces the most natural European-language translations, and the document mode keeps formatting intact.
OpenAI's video model with a real storyboard editor
Sora generates up to 20-second clips with convincing physics, and its storyboard timeline lets you direct shots instead
The open model line that made photoreal images cheap
Black Forest Labs' FLUX family sets the bar for prompt adherence and human anatomy, and the smaller variants run on a
The AI-first code editor
Built on VS Code, Cursor offers codebase-aware chat, multi-file edits and fast autocomplete for real engineering work.
An AI assistant that's helpful, harmless and honest
Anthropic's Claude excels at nuanced reasoning, long-document analysis and careful writing. Popular with professionals
The conversational AI that started it all
OpenAI's flagship chatbot for writing, coding, brainstorming and analysis. The free tier now includes GPT-4o mini, with
AI-text detector built for long documents
Pangram scores whole documents rather than single paragraphs, which makes it far less trigger-happy than sentence-level
One API key for hundreds of models
OpenRouter proxies 400+ models behind an OpenAI-compatible API, with automatic fallbacks, price comparison and per-key
Visual workflow automation you can self-host
n8n connects 400+ apps in a node editor, adds AI agent steps, and can run entirely on your own server.
Chat with your own documents, then hear the summary
NotebookLM — now branded Gemini Notebook — grounds every answer in the sources you upload, cites the exact passage, and
Anthropic's terminal agent that edits whole repos
Claude Code runs in your shell, reads the repository, plans a change and applies multi-file edits with tests.
Google's image model that keeps a character consistent
Nano Banana (Gemini 2.5 Flash Image) edits photos conversationally and holds a subject's identity across many generated
Andrew Ng's short courses on applied AI
DeepLearning.AI offers short, practical courses on prompting, RAG, agents and fine-tuning, many free with a certificate.
Run open models locally with one command
Ollama downloads, runs and serves open models on your own machine with a single command and an OpenAI-compatible API.
Kuaishou's video model, now with a global site
Kling (可灵) is the strongest Chinese video model for human motion, and the English site adds lip-sync, motion brushes
Google's cinematic video model with native audio
Veo produces 8-second shots with synced sound, camera-direction prompts and reference-image control, available through
A general agent that finishes the task, not just answers
Manus runs a cloud sandbox where it browses, writes files, builds slides and ships a finished deliverable while you
Alibaba's multilingual assistant, free and open-weight
Alibaba's Qwen family is the most-downloaded open model line in the world; Qwen Chat is the hosted English interface
The platform for open machine learning
Host, discover and run open models and datasets. The GitHub of ML, with Spaces for demos.
Text-to-image known for artistic quality
Midjourney produces highly stylized, painterly images and remains a favorite for concept art and illustration.
The AI answer engine that shows its sources
Perplexity answers questions with cited web results, making it a research-first alternative to traditional search.
ByteDance's consumer front end for Seedream images and Seedance video
Jimeng is the Chinese consumer app built on ByteDance's Seedream image and Seedance video models. It is where most of
Long-form fiction with a codex that remembers canon
NovelCrafter is a writing environment for novels: a codex of characters and places feeds the model so drafts stay
Search filings, transcripts and research in one place
AlphaSense indexes broker research, filings, transcripts and news, then answers questions with citations for investment
Full-stack SEO and competitive intelligence
Semrush combines keyword research, site audits, backlink analysis and competitor ad tracking, with AI assistance across
University courses and certificates online
Coursera hosts courses from universities and companies, with graded assignments, certificates and full degree paths.
Interactive courses in maths, science and CS
Brilliant teaches through short interactive problems rather than videos, covering maths, logic, data and programming.
The reference wiki for prompting techniques
Maintained by DAIR.AI, this guide documents prompting techniques with papers, examples and papers-backed explanations.
Free, structured course on prompt engineering
Learn Prompting is an open course that walks from basic prompting to chain-of-thought, RAG and agent patterns, with
Free Gemini prototyping with a generous API tier
AI Studio lets you test Gemini models, tune prompts and generate code, with a free API tier that is unusually usable.
Support AI that actually resolves tickets
Fin answers customer questions from your help centre and can perform actions like refunds, with a resolution rate you
Enrich any list of people or companies, then personalise
Clay waterfalls through 100+ data providers to enrich leads, then uses AI to write personalised outreach per row.
Bilingual web pages and PDFs side by side
Immersive Translate renders a second language under every paragraph on a page or PDF, using whichever engine you prefer.
Free scholarly search with citation context
Semantic Scholar indexes 200M+ papers and shows what citing papers actually say about each reference — free and