# StackPick

> An opinionated directory and ranking of AI tools for vibe coders and indie builders — which tool to use for which job, and what shipped this week. Each score is a weighted editorial estimate: Ease 22%, Power 33%, Price (indie value) 15%, Vibe-coder fit 30%. Price means value for an indie budget, not the sticker price. Scores are not measured benchmarks. Affiliate links may be used to monetize, but they never affect scores or rankings, and rankings are never sponsored.

Rankings as of Oct 19, 2026.

## Categories

- [AI code editors](https://stackpick.app/category/ai-code-editors): The editor you live in. #1 Cursor, editorial score 8.6. Rankings as of Oct 15, 2026.
- [AI app builders](https://stackpick.app/category/ai-app-builders): Prompt to a running app. #1 Lovable, editorial score 8.1. Rankings as of Oct 15, 2026.
- [Coding and CLI agents](https://stackpick.app/category/coding-agents): An agent that edits the repo. #1 Claude Code, editorial score 8.3. Rankings as of Oct 15, 2026.
- [Agent frameworks](https://stackpick.app/category/agent-frameworks): The harness around a model. #1 Vercel AI SDK, editorial score 8.5. Rankings as of Oct 15, 2026.
- [LLM APIs and model providers](https://stackpick.app/category/llm-apis): The model behind the feature. #1 Anthropic API, editorial score 8.3. Rankings as of Oct 15, 2026.
- [Hosting and deploy](https://stackpick.app/category/hosting): Where the app stays up. #1 Vercel, editorial score 8.2. Rankings as of Oct 15, 2026.
- [Databases and backend](https://stackpick.app/category/databases): Somewhere for the app to remember. #1 Supabase, editorial score 8.5. Rankings as of Oct 19, 2026.
- [Auth](https://stackpick.app/category/auth): Sign-in without a week of edge cases. #1 Clerk, editorial score 8.2. Rankings as of Oct 15, 2026.
- [Payments](https://stackpick.app/category/payments): Get paid. #1 Polar, editorial score 8.0. Rankings as of Oct 15, 2026.
- [Design to code](https://stackpick.app/category/design-to-code): From a picture to components. #1 Google Stitch, editorial score 7.3. Rankings as of Oct 15, 2026.
- [Testing and debugging](https://stackpick.app/category/testing): Catch the break before your users do. #1 Sentry, editorial score 7.7. Rankings as of Oct 15, 2026.

## Stacks

- [Ship a SaaS this weekend](https://stackpick.app/stack/ship-a-saas-weekend): Editor, UI, Postgres, payments, a URL, and somewhere to see the crash.
- [Build an AI agent](https://stackpick.app/stack/build-an-ai-agent): A TypeScript harness, a model API, a place to store state, and an editor that can keep up.
- [Prompt an app into existence](https://stackpick.app/stack/prompt-to-app): Stay inside a generator until the demo works. Leave when the diff matters.
- [Design to deploy](https://stackpick.app/stack/design-to-deploy): A design file, a code generator, an editor to finish it, then a host.

## Guides

- [Cursor vs Claude Code](https://stackpick.app/compare/cursor-vs-claude-code): Cursor scores 8.6 and Claude Code 8.3 on StackPick. Where each wins, how they differ, and why many indie builders run both.
- [Lovable vs Bolt vs v0](https://stackpick.app/compare/lovable-vs-bolt-vs-v0): Lovable 8.1, Bolt 7.9, v0 7.5 on StackPick. Full-stack app, browser sandbox, or React UI: pick by the job, not the hype.
- [The best AI code editor in 2026](https://stackpick.app/best/ai-code-editor): Cursor 8.6, GitHub Copilot 7.8, Zed 7.5, Devin Desktop 7.3, Continue 7.1, Augment Code 7.0, JetBrains AI Assistant 6.3. One pick per need.
- [The best AI app builder in 2026](https://stackpick.app/best/ai-app-builder): Lovable leads at 8.1. Bolt is next at 7.9. Ranked AI app builders for vibe coders. Scores are editorial estimates.
- [The best AI coding agent in 2026](https://stackpick.app/best/ai-coding-agent): Claude Code leads at 8.3. Cline is next at 7.6. Ranked coding and CLI agents for vibe coders. Scores are editorial estimates.

## Tools

- [Cursor](https://stackpick.app/tool/cursor): AI code editors. Editorial score 8.6. AI editor on a VS Code fork, with tab, inline edit, and an agent.
- [GitHub Copilot](https://stackpick.app/tool/github-copilot): AI code editors. Editorial score 7.8. GitHub's AI pair programmer across VS Code, JetBrains, and its own app.
- [Zed](https://stackpick.app/tool/zed): AI code editors. Editorial score 7.5. A fast collaborative editor with built-in edit predictions.
- [Devin Desktop](https://stackpick.app/tool/devin-desktop): AI code editors. Editorial score 7.3. Also known as Windsurf. The Windsurf IDE, renamed, with an agent command center in front.
- [Continue](https://stackpick.app/tool/continue): AI code editors. Editorial score 7.1. Open-source autopilot for VS Code and JetBrains. Bring your own model.
- [Augment Code](https://stackpick.app/tool/augment-code): AI code editors. Editorial score 7.0. AI coding assistant pitched at large, existing codebases.
- [JetBrains AI Assistant](https://stackpick.app/tool/jetbrains-ai): AI code editors. Editorial score 6.3. AI help inside IntelliJ, PyCharm, WebStorm, and the rest of the suite.
- [Lovable](https://stackpick.app/tool/lovable): AI app builders. Editorial score 8.1. Also known as GPT Engineer. Describe a product. Lovable builds a full-stack app you can keep editing in chat.
- [Bolt](https://stackpick.app/tool/bolt): AI app builders. Editorial score 7.9. Also known as Bolt.new. StackBlitz's in-browser builder. Prompt a full-stack app and run it there.
- [Replit](https://stackpick.app/tool/replit): AI app builders. Editorial score 7.7. Browser IDE plus an agent that can build and host the app.
- [v0](https://stackpick.app/tool/v0): AI app builders. Editorial score 7.5. Vercel's generator for React and Tailwind UI from a prompt.
- [Base44](https://stackpick.app/tool/base44): AI app builders. Editorial score 7.0. Prompt-to-app builder for internal tools and simple products.
- [Firebase Studio](https://stackpick.app/tool/firebase-studio): AI app builders. Editorial score 6.8. Google's cloud workspace for building apps with Firebase nearby.
- [Claude Code](https://stackpick.app/tool/claude-code): Coding and CLI agents. Editorial score 8.3. Anthropic's coding agent for the terminal and supported editors.
- [Cline](https://stackpick.app/tool/cline): Coding and CLI agents. Editorial score 7.6. Also known as Claude Dev. Open-source coding agent inside VS Code. Bring your own model.
- [Codex CLI](https://stackpick.app/tool/codex): Coding and CLI agents. Editorial score 7.5. Also known as OpenAI Codex CLI. OpenAI's open-source terminal agent for editing a repo.
- [Aider](https://stackpick.app/tool/aider): Coding and CLI agents. Editorial score 7.4. Open-source pair programming in the terminal. Git commits included.
- [OpenCode](https://stackpick.app/tool/opencode): Coding and CLI agents. Editorial score 7.4. Open-source coding agent you run in the terminal.
- [Gemini CLI](https://stackpick.app/tool/gemini-cli): Coding and CLI agents. Editorial score 7.2. Google's open-source terminal agent, wired to Gemini.
- [Devin](https://stackpick.app/tool/devin): Coding and CLI agents. Editorial score 6.9. Cognition's cloud agent that takes a task and works on its own machine.
- [Vercel AI SDK](https://stackpick.app/tool/vercel-ai-sdk): Agent frameworks. Editorial score 8.5. Also known as AI SDK. TypeScript toolkit for streaming, tool calls, and chat UI.
- [Mastra](https://stackpick.app/tool/mastra): Agent frameworks. Editorial score 8.0. TypeScript framework for agents, workflows, and evals.
- [CrewAI](https://stackpick.app/tool/crewai): Agent frameworks. Editorial score 7.7. Multi-agent crews: roles, tasks, and a Python API.
- [Pydantic AI](https://stackpick.app/tool/pydantic-ai): Agent frameworks. Editorial score 7.6. Python agent framework from the Pydantic team, typed end to end.
- [LangGraph](https://stackpick.app/tool/langgraph): Agent frameworks. Editorial score 7.3. Graph-shaped agents with state, from the LangChain team.
- [LlamaIndex](https://stackpick.app/tool/llamaindex): Agent frameworks. Editorial score 7.0. Also known as GPT Index. A data framework for connecting documents and models.
- [LangChain](https://stackpick.app/tool/langchain): Agent frameworks. Editorial score 6.7. The original LLM application framework, plus a hosted LangSmith side.
- [Anthropic API](https://stackpick.app/tool/anthropic): LLM APIs and model providers. Editorial score 8.3. Claude models via API. The pair to Claude Code.
- [OpenAI API](https://stackpick.app/tool/openai): LLM APIs and model providers. Editorial score 8.0. The default model API. GPT models, tools, and a platform console.
- [OpenRouter](https://stackpick.app/tool/openrouter): LLM APIs and model providers. Editorial score 7.9. One API key for many models, including labs and open weights.
- [Groq](https://stackpick.app/tool/groq): LLM APIs and model providers. Editorial score 7.7. Fast inference API for open models.
- [Gemini API](https://stackpick.app/tool/google-gemini): LLM APIs and model providers. Editorial score 7.6. Google's Gemini models, with a free tier that moves.
- [Mistral](https://stackpick.app/tool/mistral): LLM APIs and model providers. Editorial score 6.7. European model lab with an API and open-weight releases.
- [Vercel](https://stackpick.app/tool/vercel): Hosting and deploy. Editorial score 8.2. Push a frontend or Next.js app and get a URL.
- [Cloudflare Workers](https://stackpick.app/tool/cloudflare): Hosting and deploy. Editorial score 7.8. Run code on Cloudflare's edge, plus Pages for sites.
- [Railway](https://stackpick.app/tool/railway): Hosting and deploy. Editorial score 7.4. Deploy a service from a repo without writing a platform file first.
- [Netlify](https://stackpick.app/tool/netlify): Hosting and deploy. Editorial score 7.2. Git-based hosting for web apps, with functions and forms.
- [Render](https://stackpick.app/tool/render): Hosting and deploy. Editorial score 7.2. Web services, static sites, cron, and databases on one platform.
- [Supabase](https://stackpick.app/tool/supabase): Databases and backend. Editorial score 8.5. Postgres, auth, storage, and realtime in one project.
- [Neon](https://stackpick.app/tool/neon): Databases and backend. Editorial score 8.0. Serverless Postgres that branches like a git repo.
- [Convex](https://stackpick.app/tool/convex): Databases and backend. Editorial score 7.9. A reactive backend: queries, mutations, and functions in TypeScript.
- [Firebase](https://stackpick.app/tool/firebase): Databases and backend. Editorial score 7.4. Google's backend platform: Auth, Firestore, storage, and hosting.
- [Upstash](https://stackpick.app/tool/upstash): Databases and backend. Editorial score 7.4. Serverless Redis and Kafka with a per-request price.
- [Wix Headless](https://stackpick.app/tool/wix-headless): Databases and backend. Editorial score 7.2. Wix CMS, commerce, bookings, and member login for a frontend you build.
- [Clerk](https://stackpick.app/tool/clerk): Auth. Editorial score 8.2. Hosted auth with prebuilt components for Next.js and friends.
- [Auth.js](https://stackpick.app/tool/authjs): Auth. Editorial score 6.9. Also known as NextAuth. Open-source auth for Next.js and other frameworks. You host it.
- [WorkOS](https://stackpick.app/tool/workos): Auth. Editorial score 6.9. Enterprise SSO, directory sync, and auth for B2B apps.
- [Polar](https://stackpick.app/tool/polar): Payments. Editorial score 8.0. Billing for developers, with an open-source core.
- [Stripe](https://stackpick.app/tool/stripe): Payments. Editorial score 8.0. Cards, subscriptions, and a lot of billing API.
- [Lemon Squeezy](https://stackpick.app/tool/lemonsqueezy): Payments. Editorial score 7.4. A merchant of record for software sales.
- [Google Stitch](https://stackpick.app/tool/google-stitch): Design to code. Editorial score 7.3. Google Labs tool that turns a prompt into a UI and frontend code.
- [Figma](https://stackpick.app/tool/figma): Design to code. Editorial score 7.2. The design tool. Dev Mode is how engineering reads the file.
- [Builder.io](https://stackpick.app/tool/builder): Design to code. Editorial score 6.9. Visual CMS and design-to-code that emits framework components.
- [Sentry](https://stackpick.app/tool/sentry): Testing and debugging. Editorial score 7.7. Error monitoring. The stack trace from production, with the release.
- [Playwright](https://stackpick.app/tool/playwright): Testing and debugging. Editorial score 7.6. Microsoft's open-source framework for end-to-end browser tests.
- [Jam](https://stackpick.app/tool/jam): Testing and debugging. Editorial score 7.4. Bug reports with console, network, and a replay, not a vague screenshot.
- [Checkly](https://stackpick.app/tool/checkly): Testing and debugging. Editorial score 7.1. Monitoring that runs Playwright-style checks against production.

## Site

- [How scores work](https://stackpick.app/about): Method, weights, and the affiliate policy. Affiliate links may be used to monetize, but they never affect scores or rankings, and rankings are never sponsored.
- [Submit a tool](https://stackpick.app/submit): Submit an AI dev tool for review. Submissions are not published automatically.
- [Advertise](https://stackpick.app/advertise): Featured listing, newsletter sponsor slot, or priority review. Paid placement is labeled and does not change a score.
- [New this week](https://stackpick.app/feed.xml): RSS feed of releases inside the current seven-day window.
- [Full tool text](https://stackpick.app/llms-full.txt): Description, best for, pros, cons, and pricing summary for every tool.