✍️ AI Writing
See all writing tools →
💻 AI Coding
See all coding tools →
💻
Cursor
AI Coding
8.8
out of 10

The Cursor 3 rebuild puts coding agents — not autocomplete — at the center: Composer 2.5, parallel agents, and multi-repo workspaces. Category-leading on features; the usage-based credit pricing is the trade-off.

Code editorAutocompleteFree plan
🐙
GitHub Copilot
AI Coding
8.9
out of 10

The best value AI coding tool. Works in every major IDE at $10/month — no switching editors required.

Multi-IDEFree plan$10/mo
🌊
Windsurf (Devin Desktop)
AI Coding
8.2
out of 10

Windsurf, rebranded Devin Desktop by Cognition in June 2026. A VS Code-based IDE betting on an open multi-agent layer (ACP) as Cascade is retired.

AgenticFree planOpen ACP
Bolt.new
AI Coding
8.3
out of 10

Fastest path from idea to live app. Build full-stack MVPs from a single prompt directly in your browser.

No-codeFull-stackPrototyping
🔁
Replit
AI Coding
8.3
out of 10

The best browser-based AI coding environment. Replit Agent builds full apps from plain-language prompts with zero local setup required.

Browser-basedAI AgentFree plan
💗
Lovable
AI Coding
8.1
out of 10

The full-stack app builder leading the "vibe coding" category. One prompt generates a React + Supabase app and deploys it — but the credit system is opaque.

No-codeFull-stackSupabase
v0 by Vercel
AI Coding
8.4
out of 10

Among the strongest UI and React/Next.js generation in the category, with one-click Vercel deploy. The Feb 2026 rebuild added Git and database connectivity — but it stays frontend-first.

UI generationNext.jsVercel deploy
🛸
Google Antigravity 2.0
AI Coding
7.5
out of 10

Google's agent-first development platform. Parallel multi-agent execution and deep Google ecosystem integration — undermined by a rocky 2.0 launch and a CLI that wasn't shippable on day one.

Multi-agentGemini 3.5Free plan
ZCode
AI Coding
7.8
out of 10

Z.ai's free agent-first "Agentic Development Environment" on the open-weight GLM-5.2 — credibly the strongest open-weight coding model on Z.ai-published benchmarks (62.1% SWE-bench Pro), MIT weights, 1M context. The real decision is governance: Z.ai's US Entity List status and the National Intelligence Law's reach over its cloud API make self-hosting vs. hosted-API routing the load-bearing call. Includes a conflict-of-interest disclosure.

Open weightsAgent-first ADEGLM-5.2
🐋
DeepSeek V4
AI Coding
8.1
out of 10

Open-weight LLM scoring 80.6% on SWE-bench at $0.435/M input — a fraction of the cost of Claude Opus 4.8. Strong open-weight coding benchmarks at a developer-API price, with real data-residency caveats.

Open weights1M contextOpenAI-compatible
✳️
Claude Code
AI Coding
8.0
out of 10

Anthropic's coding agent — 1M context, Opus 4.8 at 88.6% SWE-bench, and Agent Teams running parallel instances on git worktrees. The strongest agent for senior developers; team pricing is the catch. Includes a full conflict-of-interest disclosure.

Agent Teams1M contextOpus 4.8
🌙
Kimi Code
AI Coding
7.8
out of 10

Moonshot AI's open-source Claude Code alternative. Kimi K2.6 scores 80.2% on SWE-bench at $0.60/M input — roughly 8x cheaper than Claude Opus 4.8 — and runs every Claude Code MCP server unchanged. Agent Swarms coordinate up to 300 parallel agents. CLI-only, with the same data-residency caveats as DeepSeek.

Open weightsMCP compatibleAgent Swarms
🛠️
Grok Build
AI Coding
6.7
out of 10

xAI's terminal-native coding agent, and the third lab-backed entry alongside Claude Code and Codex CLI. Its 8 parallel sub-agents in isolated Git worktrees are a genuine architectural first — but grok-build-0.1 scores 70.8% SWE-Bench (vs 88.6% for Claude Code), Arena Mode isn't live yet, and $99/mo intro pricing reverts to $299/mo after 6 months. An early beta worth watching.

Git worktreesMCP compatibleEarly beta
🜂
MiniMax M3
AI Coding
7.9
out of 10

The first open-weight model to combine frontier coding (59% SWE-Bench Pro), a 1M-token context, and native text/image/video input. The new MSA sparse-attention architecture runs 15.6x faster at 1M context, and $0.60/M input is ~8x cheaper than Claude Opus 4.8. The catches: vendor-reported benchmarks awaiting independent verification, weights not shipped at launch, and the same data-residency questions as DeepSeek and Kimi.

Open weightsMultimodal1M context
⌨️
OpenCode
AI Coding
8.4
out of 10

Open-source, MIT-licensed, model-agnostic coding agent that entered LogRocket's June 2026 rankings at #1, displacing Cursor. 75+ providers from one config, LSP diagnostics fed back into the agent loop, and free local/air-gapped operation. The catches: ~78% slower than Claude Code on the same model, terminal-first setup, and community-only support. Includes a conflict-of-interest disclosure.

Open sourceModel-agnosticAir-gapped
🔀
aider
AI Coding
7.8
out of 10

The git-native pioneer, reassessed. aider auto-commits every edit to git with a descriptive message — the most granular git integration among CLI coding agents — and it's free, Apache-licensed, model-agnostic, and runs fully local. The 2026 catch: development has slowed (last repo push May 22) and the curated model guidance is stale while peers like OpenCode ship weekly. Includes a conflict-of-interest disclosure.

Git-nativeOpen sourceModel-agnostic
🧠 AI Productivity
See all productivity tools →
🤖
ChatGPT
AI Productivity
9.0
out of 10

The most versatile AI assistant available. Handles writing, research, coding, analysis, and more through a simple chat interface.

General purposeFree planVersatile
📝
Notion AI
AI Productivity
8.4
out of 10

The best AI assistant for teams already in Notion. Workspace-aware Q&A surfaces anything you need from your docs instantly.

Knowledge baseAI AgentsFree plan
🤝
Claude Cowork
AI Productivity
8.2
out of 10

Autonomous background tasks and file system access, included in Claude Pro. Scheduled task runner, vertical bundles for Legal, Small Business, and Marketing Ops.

File accessScheduled tasksBundles
🖥️
Perplexity Computer
AI Productivity
7.4
out of 10

Cloud AI agent that orchestrates 19+ models in parallel to run multi-step tasks autonomously. The most capable consumer agent of 2026 — but locked to $200/month Max with opaque credits.

Multi-modelBrowser automationAI Agents
📓
NotebookLM
AI Productivity
8.7
out of 10

Google's source-grounded research assistant. Answers only from your documents, turns them into podcast-style Audio Overviews, and ships a genuinely powerful free tier.

Audio Overviews1M contextFree plan
🪄
ZoomMate
AI Productivity
7.2
out of 10

Zoom's agentic AI add-on that turns meeting conversations into finished work — not just summaries. Updates Salesforce, Jira, and Slack and builds decks from meeting context. Credible direction, but brand-new, North America only, and unproven at scale.

AgenticIntegrationsEnterprise
🦙
Ollama + Open WebUI
AI Productivity
8.4
out of 10

A free, self-hosted "private ChatGPT" you run on your own hardware. Ollama serves local models, Open WebUI is the chat UI with RAG, web search, and multi-user. Capable for the everyday tasks that don't need a frontier model — with full data ownership.

Self-hostedOpen sourceLocal LLM
🎨 AI Image
See category →
⚡ Automation
See category →
🎬 AI Video
See category →
🎞️
Grok Imagine Video 1.5
AI Video
7.5
out of 10

xAI's cheap, fast image-to-video model with native audio. Tops the image-to-video arena, but a documented content-moderation failure weighs heavily on the score.

Image-to-videoNative audioSafety caveat
🔍
Perplexity AI
Productivity
8.8
out of 10

The best AI research tool most people aren't using yet. Sourced, cited answers from the live web — more reliable than ChatGPT for facts.

Cited sourcesReal-time webResearch
✍️
Grammarly
AI Writing
8.5
out of 10

The gold standard for AI-assisted editing. Works everywhere you write, catches what re-reads miss, and the annual plan is excellent value.

GrammarTone detectionPlagiarism check
🎙️
Descript
AI Video
7.9
out of 10

Text-based video editing for spoken-word creators. Overdub voice cloning, Studio Sound, and filler word removal in one platform.

PodcastingFree planOverdub
🎙️
ElevenLabs
AI Video
8.3
out of 10

The AI voice-quality leader — text-to-speech, cloning, and dubbing. The honest catch: the top Eleven v3 model isn't built for real-time, and credit-based pricing is harder to predict than the sticker price.

Text-to-speechVoice cloningFree plan
🎬
Runway
AI Video
8.9
out of 10

The benchmark for AI video generation. Gen-4.5 delivers cinematic quality with strong character consistency across shots.

Text-to-videoGen-4.5Cinematic
🎭
Synthesia
AI Video
8.0
out of 10

The leading AI avatar platform for business. Create multilingual training videos with 230+ avatars — no camera or studio needed.

AI avatarsMultilingualTraining video
🎥
HeyGen
AI Video
8.1
out of 10

The realism leader in AI avatar video. Avatar IV, photo-to-avatar creation, and 175+ language translation built for marketing and sales.

Avatar IVPhoto-to-avatar175+ languages
🔜Reviews coming soon — we're testing Sora, Kling, and more.