# StackPick

> StackPick (usestackpick.com) is an opinionated AI tool directory for vibe coders and builders. For each job (coding agents, app builders, databases, hosting, AI APIs, agent frameworks) it ranks a few tools, gives an honest one-line verdict and current pricing, and links to the official site. A weekly feed covers new AI dev tool releases with sources.

Last updated: 2026-10-19. Pricing changes often; verify on the official pricing page before buying. Sponsored placements are labeled and do not change rankings.

## Pages

- [Tool directory](https://usestackpick.com/): ranked picks by use case
- [New this week](https://usestackpick.com/new): recent AI dev tool releases with sources
- [Get featured](https://usestackpick.com/featured): sponsorship options for tool makers
- [Raw data: tools](https://usestackpick.com/data/tools.json)
- [Raw data: releases](https://usestackpick.com/data/news.json)

## Coding agents & IDEs

Where you actually write (or direct) the code.

1. [Cursor](https://cursor.com): best for Best all-round AI IDE. AI editor with agents, cloud agents and Bugbot code review. Plans are now bundled with Grok/X subscriptions; third-party models (Claude, GPT) cost extra at API list price. Pricing: Free; Plus $8/mo (Grok models only); Premium $30/mo; Super $100/mo; Ultra $200/mo; Teams from $30/user/mo
2. [Claude Code](https://claude.com/product/claude-code): best for Best for terminal-first agentic work. Agent that lives in your terminal and runs long multi-step tasks. Heavy users tend to need a Max plan. Pricing: Included with Claude Pro ($20/mo, or $17/mo billed annually); Max from $100/mo; or pay-as-you-go API
3. [GitHub Copilot](https://github.com/features/copilot): best for Best budget pick / GitHub teams. Cheapest mainstream paid option with deep GitHub integration and many models. Chat/agent use is metered by AI Credits. Pricing: Free; Pro $10/mo, Pro+ $39/mo, Max $100/mo; Business/Enterprise: see site
4. [Windsurf (Devin Desktop)](https://devin.ai/desktop): best for Best Cursor alternative. Windsurf was renamed Devin Desktop by Cognition on June 2, 2026: same IDE, now with an agent command center. Daily/weekly quotas; extra usage at API pricing. Pricing: Free; Pro $20/mo; Max $200/mo; Teams $80/mo + $40/mo per full seat
5. [OpenCode](https://opencode.ai): best for Best open-source agent. Open-source terminal/desktop coding agent; bring your own model. 2.0 rebuilt it around a shared local service with Code Mode and a new SDK. Pricing: Open source (you pay your model provider)
6. [Harness](https://www.autonomous.ai/harness): best for Best multi-agent command center. Open-source Mac/Linux app that runs Claude Code, Codex, Cursor, OpenCode and more side by side across machines, with mobile apps for voice replies. Agents keep working after you close the desktop app. Pricing: Free and open source (MIT)

## App builders

Prompt to working app, little or no code.

1. [Lovable](https://lovable.dev): best for Best for beginners & small teams. Polished prompt-to-app with built-in Cloud backend. Credits are shared by unlimited workspace members. Pricing: Free (5 credits/day, up to 30/mo); Pro $25/mo; Business from $50/mo
2. [Bolt](https://bolt.new): best for Best for heavy solo builders. In-browser full-stack builder metered by tokens. Generous free daily allowance. Pricing: Free (300K tokens/day, 1M/mo); Pro $25/mo; Teams $30/member
3. [v0](https://v0.app): best for Best for React/Next.js UI on Vercel. Vercel's builder; strongest at front-end and ships straight to Vercel. Billed per user. Pricing: Free (7 messages/day); Plus $30/user/mo; Business $100/user/mo
4. [Replit](https://replit.com): best for Best for a real IDE + hosting in one. Agent plus full cloud IDE, database and deploys. Good when you outgrow pure prompting. Pricing: Free Starter; Core $20/mo ($18 annual); Pro $100/mo ($90 annual)

## Databases & backend

Auth, data and storage for what you build.

1. [Supabase](https://supabase.com): best for Best default backend. Postgres plus auth, storage and realtime; the most common backend in vibe-coded apps. Free projects pause after a week idle. Pricing: Free (500 MB DB, 2 active projects, pauses after 1 week idle); Pro from $25/mo incl. one Micro project, extra projects add compute; Team from $599/mo
2. [Firebase](https://firebase.google.com): best for Best for mobile apps. Google's mature BaaS with auth, Firestore and push messaging. NoSQL model differs from Postgres-based options. Pricing: Spark plan free; Blaze pay-as-you-go
3. [Neon](https://neon.com): best for Best serverless Postgres. Postgres that scales to zero with instant branching, handy for preview environments and agents. Pricing: Free plan; Launch and Scale are usage-based
4. [Convex](https://convex.dev): best for Best for realtime TypeScript apps. Reactive backend where queries and functions are plain TypeScript. Great DX for realtime apps. Pricing: Free & Starter (pay as you go); Professional $25/developer/mo; Business from $2,500/mo minimum

## Hosting

Get it on the internet.

1. [Vercel](https://vercel.com): best for Best for Next.js / frontend. Best-in-class Next.js hosting with previews. Hobby plan is non-commercial only. Pricing: Hobby free (non-commercial); Pro $20/mo incl. $20 usage credit
2. [Netlify](https://www.netlify.com): best for Best for static sites on a budget. Simple Git-based deploys with a credit system covering bandwidth, compute and builds. Pricing: Free (300 credits); Personal $9/mo; Pro $20/mo
3. [Cloudflare](https://pages.cloudflare.com): best for Best for scale at low cost. Pages/Workers on a global edge network; very cheap for static and edge workloads. Pricing: Free static hosting on Pages; Workers paid usage: see site
4. [Railway](https://railway.com): best for Best for backends & long-running services. Deploy containers, workers and databases without DevOps. Good when you need a server, not functions. Pricing: Free ($1/mo credit) & 30-day $5 trial; Hobby $5/mo; Pro $20/mo per workspace; plus per-second usage

## AI APIs

Models to power the features in your app.

1. [OpenAI API](https://platform.openai.com): best for Best general-purpose default. Broadest ecosystem, SDKs and docs. Safe first choice for most features. Pricing: Pay-per-token; see pricing page
2. [Anthropic API](https://www.anthropic.com/api): best for Best for coding & agents. Claude models are a top choice for code generation and tool-using agents. Pricing: Pay-per-token; see pricing page
3. [Google Gemini API](https://ai.google.dev): best for Best for long context & multimodal. Long context windows and native multimodal input; free tier via AI Studio. Pricing: Free tier and pay-per-token; see site
4. [OpenRouter](https://openrouter.ai): best for Best for trying many models. One API key and OpenAI-compatible endpoint for hundreds of models; easy model switching. Pricing: Pay-per-token at provider rates; see site
5. [Groq](https://groq.com): best for Best for low latency. Very fast inference for open-weight models; now also hosts Meta's Muse Spark Open. Pricing: Free tier and pay-per-token; see site

## Agent builders

Frameworks for building your own agents.

1. [OpenAI Agents SDK](https://openai.github.io/openai-agents-python/): best for Best lightweight SDK. Minimal Python/JS SDK for tool-using agents with handoffs and tracing. Pricing: Open source (model usage billed)
2. [Claude Agent SDK](https://docs.claude.com/en/api/agent-sdk/overview): best for Best for coding-style agents. The harness behind Claude Code, exposed for your own agents. Pricing: SDK free; model usage billed
3. [LangGraph](https://www.langchain.com/langgraph): best for Best for complex stateful workflows. Graph-based orchestration with persistence; more setup, more control. Pricing: Open source; LangSmith platform paid
4. [Vercel AI SDK](https://ai-sdk.dev): best for Best for TypeScript web apps. Provider-agnostic TS toolkit for streaming chat, tools and agents in web apps. Pricing: Open source

## Recent releases

- 2026-10-16: [Reflection releases Beam open weights](https://huggingface.co/blog/reflection-beam). Reflection's first model, Beam, is now on Hugging Face under Apache 2.0: a 501B-parameter MoE (23B active) with a 1M-token context, built for coding and agentic work. It ships in BF16, FP8 and NVFP4 with day-0 support in transformers, vLLM and SGLang, plus a Rust-based harness.
- 2026-10-16: [Cursor cloud agents can build shareable canvases](https://cursor.com/changelog). Cursor cloud agents can now build interactive canvases as part of their work, and you can share a finished one with your team by sending its link. The same update put an Environments API in beta and raised cloud environments to 300 repos, with large ones loading 5x faster.
- 2026-10-16: [Cline v4.1.25 adds Cloud sessions](https://github.com/cline/cline/releases/tag/v4.1.25). Cline Cloud sessions, rolling out gradually, run a task in a hosted sandbox on a GitHub repo and branch you pick, and keep going after you close your editor. Cloud tasks show up in Recent and History with their status and notify you when they finish.
- 2026-10-16: [Harness: open-source command center for coding agents](https://www.autonomous.ai/harness). Autonomous launched Harness, a free open-source Mac/Linux app that runs Claude Code, Codex, Cursor, OpenCode and others side by side across machines, with a terminal client and iOS/Android apps for voice replies. Agents keep working after you close the app; traffic is end-to-end encrypted (MIT, github.com/autonomous-ai/openharness).
- 2026-10-15: [Claude Opus 5.5 Fast Mode leaves research preview](https://platform.claude.com/docs/en/build-with-claude/fast-mode). Anthropic took Opus 5.5 Fast Mode out of research preview: no waitlist and no beta header. Same model, up to 2.5x higher output tokens/sec at $8/$40 per MTok, with default rate limits on Start/Build/Scale tiers; also live on Claude in Microsoft Foundry (Azure-hosted).
- 2026-10-15: [DeepSeek V4.1 Flash Free in OpenCode](https://opencode.ai/v2/docs/console/models/). OpenCode made DeepSeek V4.1 Flash Free (model id deepseek-v4.1-flash-free) $0 for input, output and cached reads for a limited time on Console/Zen. Pick the Free variant in /models; the paid deepseek-v4.1-flash tier stays listed separately.
- 2026-10-15: [Meta releases Muse Spark Open weights](https://research.meta.ai/blog/muse-spark-open). Meta released an open-weight multimodal model (816B total / 42B active params, 1M-token context) aimed at coding and agentic work. It's available via Groq, OpenRouter, Together, Fireworks and others, and on Hugging Face.
- 2026-10: [OpenCode 2.0](https://anoma.ly/notes/opencode-2-0/). OpenCode 2.0 runs as one shared background service across TUI, web and desktop, with a rebuilt API/SDK and a JavaScript Code Mode. 1.0 plugins and the old server API break; upgrade with `opencode upgrade --major`.
- 2026-10-15: [GitHub: AI Scan for PRs and agentic autofix GA](https://github.blog/changelog/month/10-2026/). Code scanning's AI Scan for pull requests is now generally available, complementing CodeQL for languages it doesn't cover. Agentic autofix for code scanning alerts also reached GA.
- 2026-10-14: [GitHub Copilot HydraFusion Max](https://github.blog/changelog/month/10-2026/). HydraFusion got an updated model pool and better orchestration, plus a new Max option. Both are available in VS Code, the Copilot app and more.
- 2026-10-13: [Gemini 4 Argon in GitHub Copilot](https://github.blog/changelog/2026-10-13-gemini-4-argon-in-github-copilot/). Google's newest frontier model is now GA in Copilot for Pro+, Max, Business and Enterprise users. It's billed at provider list pricing under usage-based billing.
- 2026-10-15: [Java and Kotlin by IntelliJ IDEA 1.0 for VS Code & Cursor](https://blog.jetbrains.com/idea/2026/10/intellij-idea-s-java-and-kotlin-for-cursor-and-vs-code-hits-1-0/). JetBrains shipped a stable LSP-based extension bringing IntelliJ's Java/Kotlin intelligence to VS Code and Cursor, with dedicated Cursor Agents Window support. It now requires an IntelliJ IDEA Ultimate subscription after a 30-day trial.

## Contact

- Email: hello@usestackpick.com
- X: https://x.com/usestackpick