# StackGrade > A ranked directory of AI tools for vibe coders and builders. See which tool fits which job, and what just shipped. Grades are editorial judgments for vibe-coding workflows. They are not vendor scores, benchmark results, or vote averages. Dollar prices are omitted on purpose. Fields listed as unknown were left blank instead of guessed. Seed reviewed: 2026-10-15 Canonical site: https://www.getstackgrade.com ## Use cases - [AI coding assistants & IDEs](https://www.getstackgrade.com/categories/coding-assistants): Editors, completions, and agents that write and review code inside a real project. - [App builders](https://www.getstackgrade.com/categories/app-builders): Prompt-to-app products for when you want a working interface before you settle into an IDE. - [Deployment & hosting](https://www.getstackgrade.com/categories/deployment): Where the app actually runs, from preview URLs to a production hostname. - [Agents & automation](https://www.getstackgrade.com/categories/agents): Tools that take a goal, call other tools, and keep going across steps or apps. - [Design & UI](https://www.getstackgrade.com/categories/design): Interface design, site canvases, and generators that feed a design system. - [Databases & backend](https://www.getstackgrade.com/categories/databases): Auth, data, and storage you can stand up beside a vibe-coded app. - [Testing & debugging](https://www.getstackgrade.com/categories/testing): Finding what broke in production, and proving a fix still works. - [Media generation](https://www.getstackgrade.com/categories/media): Images, video, voice, and music for when the product needs more than UI. ## Ranked tools ### AI coding assistants & IDEs 1. [Cursor](https://www.getstackgrade.com/tools/cursor) — grade A+ (First choice for this job). The default editor when you want the agent in the repo with you. Best for: Day-to-day coding in an existing codebase. Pricing: Freemium. Site: https://cursor.com. 2. [Claude](https://www.getstackgrade.com/tools/claude) — grade A (Excellent fit). The coding agent to reach for when you want a careful reviewer and a terminal agent, not a new IDE. Best for: Agentic coding and review with Claude Code. Pricing: Freemium. Site: https://claude.ai. 3. [GitHub Copilot](https://www.getstackgrade.com/tools/github-copilot) — grade A (Excellent fit). The practical pick when you want AI inside VS Code or JetBrains and you already live on GitHub. Best for: Completions and chat in an existing IDE. Pricing: Freemium. Site: https://github.com/features/copilot. 4. [ChatGPT](https://www.getstackgrade.com/tools/chatgpt) — grade A (Excellent fit). Best as the place you plan the build and then hand software tasks to Codex. Best for: Planning a product, then delegating code to Codex. Pricing: Freemium. Site: https://chatgpt.com. 5. [Windsurf](https://www.getstackgrade.com/tools/windsurf) — grade A- (Strong fit, narrower lane). A real alternative in the AI IDE category. Re-check the name on the site before you recommend it. Best for: People comparing AI-native editors. Pricing: Pricing unknown. Site: https://windsurf.com. 6. [Cline](https://www.getstackgrade.com/tools/cline) — grade A- (Strong fit, narrower lane). The open-source agent to try before you pay for a dedicated AI IDE. Best for: Agentic edits in VS Code with your own model key. Pricing: Free. Site: https://cline.bot. 7. [Replit](https://www.getstackgrade.com/tools/replit) — grade B+ (Good alternative). A capable in-browser agent. A local AI IDE is a better daily driver for an existing repo on your machine. Best for: Coding without a local environment. Pricing: Freemium. Site: https://replit.com. 8. [Continue](https://www.getstackgrade.com/tools/continue) — grade B+ (Good alternative). The right open-source layer when you want chat and autocomplete on a model you control. Best for: Custom or local models inside an existing IDE. Pricing: Free. Site: https://continue.dev. 9. [Aider](https://www.getstackgrade.com/tools/aider) — grade B (Useful in a specific setup). A strong fit if you already live in the terminal and want the agent to touch git directly. Best for: Terminal-first coding in a git repo. Pricing: Free. Site: https://aider.chat. ### App builders 1. [Lovable](https://www.getstackgrade.com/tools/lovable) — grade A (Excellent fit). The first app builder to try when the starting point is a description, not a repo. Best for: Founders who want a working app from a prompt. Pricing: Freemium. Site: https://lovable.dev. 2. [Bolt](https://www.getstackgrade.com/tools/bolt) — grade A (Excellent fit). A top pick for a full-stack prototype that should run in the browser immediately. Best for: Full-stack prototypes without a local setup. Pricing: Freemium. Site: https://bolt.new. 3. [v0](https://www.getstackgrade.com/tools/v0) — grade A- (Strong fit, narrower lane). The app builder to open when the hard part is the interface and you are already heading to React. Best for: Interface-first React screens. Pricing: Freemium. Site: https://v0.dev. 4. [Replit](https://www.getstackgrade.com/tools/replit) — grade A- (Strong fit, narrower lane). Choose it when the app, the agent, and the URL should all live in one browser workspace. Best for: Small apps that can stay on Replit. Pricing: Freemium. Site: https://replit.com. 5. [Firebase](https://www.getstackgrade.com/tools/firebase) — grade B+ (Good alternative). Firebase Studio is the builder. The rest of Firebase is the backend it is likely to sit on. Best for: Google-stack apps that need auth and data quickly. Pricing: Freemium. Site: https://firebase.google.com. ### Deployment & hosting 1. [Vercel](https://www.getstackgrade.com/tools/vercel) — grade A+ (First choice for this job). The default host when the app is Next.js and you want a preview URL on every change. Best for: Next.js apps and frontend previews. Pricing: Freemium. Site: https://vercel.com. 2. [Cloudflare](https://www.getstackgrade.com/tools/cloudflare) — grade A (Excellent fit). The host to pick for edge apps, static sites, and anything that should be cheap at the start. Best for: Edge apps and static hosting. Pricing: Freemium. Site: https://www.cloudflare.com. 3. [Railway](https://www.getstackgrade.com/tools/railway) — grade A- (Strong fit, narrower lane). Use it when you need a real service and a database, and you do not want to run the cluster yourself. Best for: Containers and small backends. Pricing: Usage-based. Site: https://railway.com. 4. [Netlify](https://www.getstackgrade.com/tools/netlify) — grade B+ (Good alternative). A solid host for a site or a simple web app. Reach for Vercel or Cloudflare when the framework or the edge runtime is the point. Best for: Jamstack sites and simple deploys. Pricing: Freemium. Site: https://www.netlify.com. 5. [Replit](https://www.getstackgrade.com/tools/replit) — grade B (Useful in a specific setup). Hosting is real, and it is attached to the workspace. Use a general host if the app should outgrow Replit. Best for: Demos that already live in Replit. Pricing: Freemium. Site: https://replit.com. ### Agents & automation 1. [Claude](https://www.getstackgrade.com/tools/claude) — grade A (Excellent fit). A first choice for multi-step coding and knowledge work you can hand to an agent. Best for: Long coding and research tasks. Pricing: Freemium. Site: https://claude.ai. 2. [n8n](https://www.getstackgrade.com/tools/n8n) — grade A (Excellent fit). The automation tool to use when you want a workflow you can see, edit, and host yourself. Best for: Connecting APIs in a workflow you control. Pricing: Freemium. Site: https://n8n.io. 3. [ChatGPT](https://www.getstackgrade.com/tools/chatgpt) — grade A- (Strong fit, narrower lane). Codex covers software tasks. It is a weaker fit when the workflow is a pile of unrelated SaaS apps. Best for: Software tasks inside the ChatGPT apps. Pricing: Freemium. Site: https://chatgpt.com. 4. [Zapier](https://www.getstackgrade.com/tools/zapier) — grade B+ (Good alternative). Use it for SaaS-to-SaaS chores. Use n8n or a coding agent when you need more control. Best for: Business automations across SaaS apps. Pricing: Freemium. Site: https://zapier.com. 5. [Cursor](https://www.getstackgrade.com/tools/cursor) — grade B+ (Good alternative). Excellent when the job is the codebase. It is not a general cross-app automation builder. Best for: Agents whose work is a repository. Pricing: Freemium. Site: https://cursor.com. 6. [Cline](https://www.getstackgrade.com/tools/cline) — grade B+ (Good alternative). A coding agent, not a business-workflow product. Best for: Editor agents with your own model key. Pricing: Free. Site: https://cline.bot. 7. [LangGraph](https://www.getstackgrade.com/tools/langgraph) — grade B+ (Good alternative). Choose the library when you are implementing an agent. Choose Claude, n8n, or Zapier when you want a product today. Best for: Teams building their own agent runtime. Pricing: Free. Site: https://www.langchain.com/langgraph. ### Design & UI 1. [Figma](https://www.getstackgrade.com/tools/figma) — grade A (Excellent fit). Still the tool to open when more than one person has to agree on the interface. Best for: Designing UI with other people. Pricing: Freemium. Site: https://www.figma.com. 2. [v0](https://www.getstackgrade.com/tools/v0) — grade A- (Strong fit, narrower lane). Generates UI code. It does not replace a shared design tool for a team. Best for: Turning a prompt into React UI. Pricing: Freemium. Site: https://v0.dev. 3. [Framer](https://www.getstackgrade.com/tools/framer) — grade A- (Strong fit, narrower lane). Use it for a marketing site you want to publish without a separate host. Use Figma when the work is an app UI. Best for: Marketing sites on a visual canvas. Pricing: Freemium. Site: https://www.framer.com. 4. [Relume](https://www.getstackgrade.com/tools/relume) — grade B+ (Good alternative). A fast way to get website structure on a page. It is not the tool for the final visual design. Best for: Early website structure and wireframes. Pricing: Freemium. Site: https://www.relume.io. ### Databases & backend 1. [Supabase](https://www.getstackgrade.com/tools/supabase) — grade A+ (First choice for this job). The default backend to put behind a new web app when you want SQL and auth without running a server. Best for: A backend next to a vibe-coded app. Pricing: Freemium. Site: https://supabase.com. 2. [Neon](https://www.getstackgrade.com/tools/neon) — grade A (Excellent fit). Pick Neon when you want Postgres itself, especially branches per preview. Pick Supabase when you also want auth and storage. Best for: Postgres that branches and scales to zero. Pricing: Freemium. Site: https://neon.tech. 3. [Firebase](https://www.getstackgrade.com/tools/firebase) — grade A- (Strong fit, narrower lane). Use it when you want Google's auth and a document database. Use Supabase or Neon when you want Postgres. Best for: Document data and mobile-style backends. Pricing: Freemium. Site: https://firebase.google.com. 4. [Convex](https://www.getstackgrade.com/tools/convex) — grade B+ (Good alternative). A good fit when you want live data and you are happy to leave SQL behind. Best for: Apps that want live queries without a database driver. Pricing: Freemium. Site: https://www.convex.dev. ### Testing & debugging 1. [Sentry](https://www.getstackgrade.com/tools/sentry) — grade A (Excellent fit). Put this on the app as soon as someone else can hit it. It is how you find out what the happy path tests missed. Best for: Production exceptions with context. Pricing: Freemium. Site: https://sentry.io. 2. [Playwright](https://www.getstackgrade.com/tools/playwright) — grade A- (Strong fit, narrower lane). Not an AI tool. It is the test runner to add so the agent has something to prove the UI still works. Best for: End-to-end checks you can re-run. Pricing: Free. Site: https://playwright.dev. 3. [Qodo](https://www.getstackgrade.com/tools/qodo) — grade B+ (Good alternative). Use it to draft tests around a change. Use Playwright to run browser flows, and Sentry to see production. Best for: Generating tests around a change. Pricing: Freemium. Site: https://www.qodo.ai. ### Media generation 1. [Midjourney](https://www.getstackgrade.com/tools/midjourney) — grade A (Excellent fit). The image tool to open when the look matters more than editing the file in a design suite. Best for: Art direction and concept images. Pricing: Paid. Site: https://www.midjourney.com. 2. [ElevenLabs](https://www.getstackgrade.com/tools/elevenlabs) — grade A (Excellent fit). The voice tool to use when a product needs speech and you do not have a recording booth. Best for: Product voice and narration. Pricing: Freemium. Site: https://elevenlabs.io. 3. [Runway](https://www.getstackgrade.com/tools/runway) — grade A- (Strong fit, narrower lane). The practical pick for a short generated video when you do not have footage. Best for: Short generated video. Pricing: Freemium. Site: https://runwayml.com. 4. [Suno](https://www.getstackgrade.com/tools/suno) — grade B+ (Good alternative). Use it for a song or a bed of music. Use ElevenLabs when you need a speaking voice. Best for: Songs and background tracks from a prompt. Pricing: Freemium. Site: https://suno.com. ## Tool notes ### Cursor Cursor is an AI-native code editor from Anysphere. Builders use it to chat with a codebase, apply inline edits, and run agents that read and change the project. Paid plans, team features, and Bugbot code review are part of the public product. Dollar prices and quotas change, so this directory does not restate them. - Vendor: Anysphere - Pricing: A free plan exists with limited agent and chat use. Paid individual, team, and enterprise plans are listed on the official pricing page. - Unknown (not invented): current dollar prices; exact request quotas ### GitHub Copilot GitHub Copilot is GitHub's coding assistant. It offers inline completions, chat, and agent mode across editors such as VS Code and JetBrains, plus GitHub surfaces. A free individual tier exists. Business and enterprise plans add admin controls. Model lists and credit allowances change; confirm them on the plans page. - Vendor: GitHub - Pricing: Copilot Free is a limited individual tier. Paid Pro, Business, and Enterprise plans are published by GitHub. - Unknown (not invented): current dollar prices; which models are in each plan this week ### Claude Claude is Anthropic's assistant. Builders use the chat app to plan, review, and draft, and Claude Code as an agent in the terminal and editor. Anthropic also ships the Claude API and managed agents. Plan inclusions move, including whether a given chat plan includes Claude Code, so confirm that on the official pricing page. - Vendor: Anthropic - Pricing: Claude chat has a free tier and paid plans. API use is billed separately. Confirm whether Claude Code is included before you rely on a specific plan. - Unknown (not invented): current dollar prices; whether Claude Code is included in each plan ### ChatGPT ChatGPT is OpenAI's assistant. The product includes general chat and Codex for software tasks in the desktop app. A free tier exists, and paid plans add more access. Model names, voice features, and Codex limits change often; the release notes and pricing page are the source of truth. - Vendor: OpenAI - Pricing: ChatGPT is free to start. Paid individual and business plans are listed on the official pricing page. API usage is a separate bill. - Unknown (not invented): current dollar prices; current default model name; Codex quota on each plan ### Windsurf Windsurf is an AI IDE associated with Codeium, aimed at agentic editing in the editor rather than a side chat. The product has been through ownership and branding changes in public reporting. Confirm the current name, owner, and plans on the official site before you standardize on it. - Vendor: Codeium - Pricing: Public plans have included a free tier and a paid editor plan. The current amounts and the brand on the pricing page were not re-verified for this seed, so the tier is marked unknown. - Unknown (not invented): current product name; current owner; current dollar prices; official pricing URL ### Continue Continue is an open-source coding assistant that runs in VS Code and JetBrains. You connect a model provider or a local model. The extension itself is free to use; model calls are billed by whoever hosts the model. - Vendor: Continue - Pricing: The editor extension is free and open source. You pay the model provider, or you run a local model. Confirm the license in the repository. - Unknown (not invented): SPDX license identifier; hosted Continue product prices ### Aider Aider is an open-source coding agent you run in the terminal. It edits a git repo and can commit the changes. You bring an API key or a local model. There is no hosted editor to sign up for. - Vendor: Aider - Pricing: The tool is free and open source. Model API usage is billed by the provider you connect. - Unknown (not invented): SPDX license identifier ### Cline Cline is an open-source agent that runs in VS Code. It can read the project, edit files, and use the terminal. You connect a model API key. The project is also published on GitHub. - Vendor: Cline - Pricing: The extension is free and open source. You pay for the model you connect. Confirm the current license in the repository. - Unknown (not invented): SPDX license identifier; official pricing URL ### Lovable Lovable is an AI app builder. You describe a product and it generates a working app you can keep editing in the browser. It is aimed at people who want a full-stack starting point without setting up a local IDE first. Plan limits are on the vendor site and were not copied here. - Vendor: Lovable - Pricing: Lovable publishes free and paid plans. Credit or message limits apply. Confirm the current numbers on lovable.dev. - Unknown (not invented): current dollar prices; official pricing URL; which backend providers a new project uses by default ### Bolt Bolt, from StackBlitz, generates and edits full-stack apps in the browser. It uses StackBlitz's in-browser environment, so a prototype can run without a local Node install. Token or message allowances are plan-dependent and were not copied here. - Vendor: StackBlitz - Pricing: Bolt has a free starting point and paid plans. Confirm current limits on bolt.new. - Unknown (not invented): current dollar prices; official pricing URL ### v0 v0 is Vercel's generative UI product. You describe an interface and it produces React code you can edit and deploy with the Vercel toolchain. It is stronger as a UI generator than as a whole-company app platform. Plan details live on Vercel and were not copied here. - Vendor: Vercel - Pricing: v0 has a free allowance and paid plans, tied to Vercel. Confirm current credits on v0.dev. - Unknown (not invented): current dollar prices; official pricing URL ### Replit Replit is a browser IDE and hosting platform. Its agent can build and modify apps inside a Replit workspace, and the same product can host the result. That makes it a fit when the project can live on Replit. It is a less natural host for an app you intend to run somewhere else. - Vendor: Replit - Pricing: Replit publishes a free tier and paid plans, including agent usage. Confirm current limits on the pricing page. - Unknown (not invented): current dollar prices; agent quota on each plan ### Vercel Vercel is a hosting platform closely tied to Next.js. It gives preview deployments for git pushes, serverless and edge functions, and a path from a framework app to a production domain. A hobby tier exists. Commercial use and higher traffic move to paid plans. - Vendor: Vercel - Pricing: Hobby is free for personal projects. Pro and Enterprise are paid. Bandwidth, build, and function limits are on the pricing page. - Unknown (not invented): current dollar prices; included bandwidth this month ### Cloudflare Cloudflare sells DNS, CDN, and application hosting. For builders, the relevant pieces are Workers and Pages: edge compute and static or full-stack hosting on Cloudflare's network. Free tiers exist for many products. Workers usage beyond the free allowance is billed. - Vendor: Cloudflare - Pricing: Many Cloudflare products have a free tier. Workers and higher support plans are paid or usage-based past the free allowance. Confirm the product you mean on the plans page. - Unknown (not invented): current Workers request allowance; which plan a given feature requires ### Railway Railway hosts apps, databases, and workers from a repo or a container. Billing is usage-based. It is a common landing spot for a side project that needs a long-running process and a database, without operating Kubernetes. - Vendor: Railway - Pricing: Railway bills from usage. A trial credit or free allowance has existed; confirm what a new account receives on the pricing page. Dollar rates were not copied. - Unknown (not invented): current trial credit amount; per-resource rates ### Netlify Netlify hosts static sites and web apps from git, with preview deploys, forms, and serverless functions. It is a straightforward host for a marketing site or a Jamstack app. A free tier exists. - Vendor: Netlify - Pricing: A free starter plan exists. Paid plans raise limits for bandwidth, builds, and team features. Confirm numbers on the pricing page. - Unknown (not invented): current dollar prices; included build minutes ### n8n n8n is a workflow automation product. You connect triggers and actions across APIs on a visual canvas. You can use n8n Cloud or host it yourself. The source is public under a sustainable-use license, which is not the same as a standard open-source license. Read the license before you fork or resell it. - Vendor: n8n - Pricing: Cloud plans are paid. Self-hosting is available under the vendor license. Confirm both the price and the license on n8n.io. - Unknown (not invented): current cloud prices; exact license name to cite ### Zapier Zapier connects thousands of SaaS apps with trigger-and-action workflows called Zaps. It is the low-ops choice for business automation. A free plan exists and is limited by tasks. AI features have been added on top of the classic automation product. - Vendor: Zapier - Pricing: A free plan exists with task limits. Paid plans raise those limits. Confirm the current task caps on the pricing page. - Unknown (not invented): current dollar prices; free-plan task cap ### LangGraph LangGraph is an open-source library from LangChain for building stateful, multi-step agents as graphs. It is a framework you embed in your own app. LangChain also sells hosted products such as LangSmith; those are separate from the library. - Vendor: LangChain - Pricing: The LangGraph library is free to use. Hosted LangChain products have their own plans. Confirm the library license in the repository. - Unknown (not invented): SPDX license identifier; LangSmith price ### Figma Figma is the standard collaborative interface design tool. Teams design, prototype, and hand off UI in a shared file. Figma has added generative features over time. Which AI features are in a given plan changes, so confirm that on the pricing page rather than assuming every seat includes them. - Vendor: Figma - Pricing: A free starter plan exists. Professional and organization plans are paid. AI feature packaging was not copied into this seed. - Unknown (not invented): which AI features are in the free plan; current dollar prices ### Framer Framer is a visual site builder. You design on a canvas and publish to the web, with a CMS and localization on higher plans. AI layout and copy features have been part of the product. It is for marketing sites and landing pages more than for application UI systems. - Vendor: Framer - Pricing: Framer has a free plan and paid site plans. Custom domains and CMS limits depend on the plan. Dollar amounts were not copied. - Unknown (not invented): current dollar prices; official pricing URL; which AI features are in the free plan ### Relume Relume turns a site brief into a sitemap and wireframes, then exports toward Figma or Webflow. It is an early-structure tool for marketing sites, not a general image generator or a full design suite. - Vendor: Relume - Pricing: Relume publishes free and paid plans. Component and project limits depend on the plan. Confirm them on relume.io. - Unknown (not invented): current dollar prices; official pricing URL ### Supabase Supabase is a hosted Postgres platform with authentication, file storage, auto-generated APIs, and edge functions. It is the usual backend recommendation next to a vibe-coded web app. A free project tier exists. The database is Postgres, so you are not locked into a proprietary query language. - Vendor: Supabase - Pricing: A free tier exists for development projects. Pro and higher plans are paid. Compute, storage, and auth limits are on the pricing page. - Unknown (not invented): current dollar prices; whether the free tier pauses inactive projects right now ### Neon Neon is serverless Postgres. It separates storage and compute, can scale to zero, and can branch a database the way you branch git. It is a database, not an auth-and-storage suite. Pair it with your own auth if you need users. - Vendor: Neon - Pricing: A free plan exists. Paid plans raise compute, storage, and branch limits. Confirm current allowances on the pricing page. - Unknown (not invented): current dollar prices; free-plan storage allowance ### Convex Convex is a backend platform with a document database, server functions, and live queries. The client updates when data changes, without you wiring a separate API and websocket. It is a different model from hosted Postgres. - Vendor: Convex - Pricing: Convex has a free developer plan and paid plans. Function calls and storage are the usual limits. Dollar amounts were not copied. - Unknown (not invented): current dollar prices; official pricing URL; whether the server runtime is open source ### Firebase Firebase is Google's app platform. Builders use Authentication, Firestore or Realtime Database, Storage, and Hosting. Firebase Studio is the newer AI app-building surface in the same family. Firestore is a document database, not Postgres. The Spark plan is free within quotas; Blaze is pay as you go. - Vendor: Google - Pricing: Spark is a no-cost quota plan. Blaze adds pay-as-you-go pricing past those quotas. Confirm which products are on Spark before you launch. - Unknown (not invented): current Spark quotas; which Firebase Studio features are free ### Sentry Sentry is error and performance monitoring. You install an SDK, and production exceptions show up with a stack trace, release, and user context. Sentry has added AI debugging features; the current feature names and which plans include them should be checked on the site. - Vendor: Sentry - Pricing: A free developer plan exists. Team and business plans are paid and metered by errors and spans. Confirm allowances on the pricing page. - Unknown (not invented): current dollar prices; which AI features are in the free plan; SDK license to cite ### Qodo Qodo, formerly Codium, makes AI tools for tests and code review. The product line includes an IDE plugin and a pull-request reviewer. It is aimed at generating and checking tests around a change, not at hosting the app or watching production errors. - Vendor: Qodo - Pricing: Qodo has published free and paid developer plans. Current prices were not copied. Confirm them on qodo.ai. - Unknown (not invented): current dollar prices; official pricing URL; which repositories the reviewer supports ### Playwright Playwright is Microsoft's open-source framework for end-to-end browser tests. It drives Chromium, Firefox, and WebKit. It is not an AI model. It is here because vibe-coded apps still need a check that the click path works. AI IDEs can write Playwright tests; Playwright is what runs them. - Vendor: Microsoft - Pricing: Playwright is free and open source. You pay only for the CI minutes or machines you run it on. - Unknown (not invented): SPDX license identifier ### Midjourney Midjourney generates images from text prompts. It is known for a polished default aesthetic and has offered a web app in addition to its earlier Discord workflow. For a long stretch it was sold as a subscription without a standing free plan. Confirm whether that is still true before you send someone there. - Vendor: Midjourney - Pricing: Midjourney has sold subscriptions rather than a permanent free tier. A trial or a free allowance may exist at times. Confirm the current plans on the official site. Dollar amounts were not copied. - Unknown (not invented): current dollar prices; whether a free trial is open; official pricing URL; current commercial-use terms ### Runway Runway is a video generation and editing product. Builders use it for short clips, image-to-video, and video edits from a prompt. Exports are credit-based on paid plans. A limited free starting point has been part of the product; confirm what it includes. - Vendor: Runway - Pricing: Runway has a free starting allowance and paid credit plans. Video seconds burn credits quickly. Confirm current plans on runwayml.com. - Unknown (not invented): current dollar prices; free credit allowance; official pricing URL ### ElevenLabs ElevenLabs generates speech from text and can clone a voice from a sample. Builders use it for product voice, narration, and prototypes of audio interfaces. A free tier exists and is capped by characters. Voice cloning has consent and plan restrictions. - Vendor: ElevenLabs - Pricing: A free character allowance exists. Paid plans raise the cap and unlock more voice features. Confirm the current character counts on the pricing page. - Unknown (not invented): current dollar prices; free-tier character cap ### Suno Suno generates music, including vocals, from a text prompt. It is aimed at songs and background tracks rather than voiceover. A free tier has existed with daily credits, and paid plans raise the allowance. Commercial rights depend on the plan. - Vendor: Suno - Pricing: A free credit allowance has been available, with paid plans for more songs and commercial use. Confirm the current rights on suno.com. Dollar amounts were not copied. - Unknown (not invented): current dollar prices; whether the free tier still includes commercial use; official pricing URL ## New releases Only releases with a public source URL and a calendar date are listed. Summaries are paraphrased. - 2026-10-15: Agentic autofix for code scanning is generally available (GitHub Copilot). GitHub changelog says agentic autofix for code scanning alerts is generally available. It explores the codebase and proposes a fix for an alert. Source: [GitHub Changelog](https://github.blog/changelog/month/10-2026/). - 2026-10-15: AI Scan for pull requests is generally available (GitHub Copilot). GitHub changelog lists code scanning AI Scan for pull requests as generally available, as a complement to CodeQL for changes in languages and frameworks CodeQL does not cover. Source: [GitHub Changelog](https://github.blog/changelog/month/10-2026/). - 2026-10-15: GPT-Live-1.1 and new voices in ChatGPT (ChatGPT). OpenAI release notes describe a GPT-Live-1.1 rollout in ChatGPT Voice, including a smaller Quip variant and additional voice options across several languages. Source: [OpenAI release notes](https://openai.com/products/release-notes/). - 2026-10-13: Gemini 4 Argon in GitHub Copilot (GitHub Copilot). GitHub changelog says Gemini 4 Argon is generally available in GitHub Copilot for complex coding, agent, and reasoning tasks. Source: [GitHub Changelog](https://github.blog/changelog/month/10-2026/). - 2026-10-09: Composer predictions in Codex (ChatGPT). OpenAI release notes describe a beta that suggests the next message in a Codex thread. Accepting a suggestion fills the box; it does not send the message on its own. Source: [OpenAI release notes](https://openai.com/products/release-notes/). - 2026-10-08: Faster steering in Codex (ChatGPT). OpenAI release notes say Codex in the ChatGPT desktop app can respond sooner when you send a follow-up to steer a running task. Source: [OpenAI release notes](https://openai.com/products/release-notes/). - 2026-10-07: Claude Haiku 5.5 on the Claude platform (Claude). Anthropic platform release notes announce Claude Haiku 5.5 for high-volume, latency-sensitive work, with a 1M token context window. Code written for Haiku 4.5 can break; read the migration notes before you switch. Source: [Claude Platform release notes](https://platform.claude.com/docs/en/release-notes/overview). - 2026-10-06: Remote control for local agents (Cursor). Cursor changelog describes viewing and messaging local agents from the Cursor iOS app. The agent keeps running on your computer, which has to stay on and online. Source: [Cursor changelog](https://cursor.com/changelog). - 2026-09-28: Claude Sonnet 5.5 on the Claude platform (Claude). Anthropic platform release notes announce Claude Sonnet 5.5 across the Claude API and several cloud platforms. The notes list request changes that can break code written for Claude Sonnet 5. Source: [Claude Platform release notes](https://platform.claude.com/docs/en/release-notes/overview). - 2026-09-23: Rollouts and Security Review (Cursor). Cursor changelog announces two bots for Teams and Enterprise: Rollouts reports deploy health per environment, and Security Review comments on exploitable bugs in a pull request. Bugbot remains the style and quality reviewer. Source: [Cursor changelog](https://cursor.com/changelog). - 2026-09-22: Claude Opus 5.5 on the Claude platform (Claude). Anthropic platform release notes announce Claude Opus 5.5 for long-running agentic coding and knowledge work, with a 1M token context window by default. The notes say thinking cannot be turned off on this model. Source: [Claude Platform release notes](https://platform.claude.com/docs/en/release-notes/overview). - 2026-09-10: Cursor Projects (Cursor). Cursor changelog describes Projects as a way to keep context on a larger body of work and delegate tasks to agents. The coordinator plans and reviews; it does not write the code itself. The beta started rolling out to all users. Source: [Cursor changelog](https://cursor.com/changelog). ## Pages - [Home](https://www.getstackgrade.com/) - [New releases](https://www.getstackgrade.com/new) - [Newsletter](https://www.getstackgrade.com/newsletter) - [Sponsor](https://www.getstackgrade.com/sponsor)