[ { "slug": "claude-code", "name": "Claude Code", "vendor": "Anthropic", "url": "https://claude.com/product/claude-code", "pricing_url": "https://claude.com/pricing", "domain": "claude.com", "status": "active", "types": [ "cli", "extension", "cloud" ], "primary": "cli", "interfaces": [ "Terminal", "VS Code / JetBrains", "Desktop app", "Web" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 20, "tiers": [ { "name": "Pro", "price": "$20/mo ($17/mo billed annually)", "note": "Includes Claude Code" }, { "name": "Max", "price": "From $100/mo", "note": "5x or 20x Pro usage" }, { "name": "Team", "price": "Per seat", "note": "For teams of 2–150" }, { "name": "Enterprise", "price": "$20/seat/mo + usage at API rates", "note": "Billed annually" }, { "name": "API (pay-as-you-go)", "price": "Per token", "note": "Via a Claude Console account" } ], "models": [ "claude" ], "models_text": "Anthropic Claude models (Opus, Sonnet, Haiku families)", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "Claude Code is Anthropic's agentic coding tool: it runs in your terminal (plus IDE, desktop and web surfaces), reads your codebase, edits files, runs commands and iterates until a task is done. It is included in paid Claude plans starting with Pro at $20/month.", "about": [ "Claude Code takes a task in plain language, explores the repository, proposes a plan, edits multiple files and runs your tests or build, then loops on failures. Because it lives in the terminal it works next to any editor and fits naturally with git, scripts and CI.", "It is powered only by Anthropic's Claude models. Usage on Pro, Max and Team plans shares one pool with Claude chat; heavy users can switch to pay-as-you-go API billing through the Claude Console. It supports MCP servers, custom slash commands, hooks and subagents for extending its toolset." ], "strengths": [ "Top-tier results on agentic terminal benchmarks (see our benchmarks page)", "Works alongside any editor; terminal-native workflow", "Strong at multi-file changes in large, unfamiliar codebases", "Extensible with MCP servers, hooks and subagents" ], "weaknesses": [ "Claude models only — no OpenAI/Gemini option", "No free tier for Claude Code itself", "Usage limits on subscription plans can bite during long sessions" ], "best_for": "Developers who are comfortable in the terminal and want an agent that can own multi-step tasks end to end.", "alternatives": [ "codex", "opencode", "aider", "amp" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=claude.com&sz=128", "yc": false, "benchmarks": [ { "name": "Terminal-Bench 4.0", "score": "64.8%", "setup": "with Claude Opus 5.5 (max effort)", "date": "Oct 2026", "source": "https://www.tbench.ai/leaderboard" }, { "name": "Terminal-Bench 2.1", "score": "89.1%", "setup": "agent-level, Artificial Analysis harness", "date": "Aug 2026", "source": "https://www.morphllm.com/ai-coding-agent" } ], "recommendable": true }, { "slug": "cursor", "name": "Cursor", "vendor": "Anysphere (part of SpaceX since Aug 2026)", "url": "https://cursor.com/", "pricing_url": "https://cursor.com/pricing", "domain": "cursor.com", "status": "active", "types": [ "ide", "cloud", "review" ], "primary": "ide", "interfaces": [ "Desktop editor (VS Code fork)", "CLI", "Web (cloud agents)" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Hobby", "price": "Free", "note": "Limited Agent requests" }, { "name": "Pro", "price": "$20/mo", "note": "Extended Agent limits, frontier models, cloud agents" }, { "name": "Pro+ / Ultra", "price": "Higher tiers", "note": "For daily and power agent users — check official site" }, { "name": "Teams", "price": "$40/user/mo", "note": "Admin, SSO, Bugbot code reviews" }, { "name": "Enterprise", "price": "Custom", "note": "Pooled usage, SCIM, audit logs" } ], "models": [ "claude", "openai", "gemini", "grok" ], "models_text": "Multiple frontier models plus Cursor's own Composer model; the lineup changes often — check the official site", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "Cursor is an AI-native code editor (a fork of VS Code) that combines fast Tab autocomplete, codebase-aware chat and an Agent that makes multi-file changes. It has a free Hobby plan; Pro costs $20/month and Teams $40/user/month.", "about": [ "Cursor keeps the familiar VS Code layout, extensions and keybindings but rebuilds the editing loop around AI: Tab predicts your next edit, the Agent plans and applies changes across files, and cloud agents can work on tasks in the background.", "Anysphere, the company behind Cursor, became a wholly owned SpaceX subsidiary on August 14, 2026. Teams get Bugbot for agentic PR code review, shared rules and skills, and admin controls. Cursor acquired Supermaven in 2024 (its tech powers Tab) and announced an agreement to acquire Graphite in December 2025." ], "strengths": [ "Best-in-class autocomplete (Tab) plus a capable Agent in one editor", "Easy switch from VS Code — extensions and settings carry over", "Multiple model providers in one subscription", "Background cloud agents and Bugbot code review" ], "weaknesses": [ "Usage-based limits can be confusing; heavy agent use costs more than $20", "Locks you into Cursor's editor (no JetBrains plugin)", "Model lineup and plans change frequently" ], "best_for": "Developers who want AI woven into everyday editing, with an agent one keystroke away.", "alternatives": [ "windsurf", "copilot", "zed", "kiro" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=cursor.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "codex", "name": "OpenAI Codex", "vendor": "OpenAI", "url": "https://chatgpt.com/codex/", "pricing_url": "https://chatgpt.com/codex/pricing/", "domain": "chatgpt.com", "status": "active", "types": [ "cli", "cloud", "extension", "review" ], "primary": "cloud", "interfaces": [ "Terminal (Codex CLI)", "IDE extension", "Web / ChatGPT", "Desktop app", "iOS" ], "open_source": true, "license": "Codex CLI is open source (Apache-2.0); cloud service proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "ChatGPT Free / Go", "price": "Included (limited)", "note": "Codex is included in all ChatGPT plans" }, { "name": "Plus", "price": "$20/mo", "note": "Codex web, CLI, IDE; code review & Slack integrations" }, { "name": "Pro", "price": "$100, $200 or $500/mo", "note": "Much higher usage; Ultrafast mode on $500" }, { "name": "Business / Enterprise", "price": "Per seat", "note": "Check official site" }, { "name": "API key", "price": "Pay per token", "note": "CLI/IDE only, no cloud features" } ], "models": [ "openai" ], "models_text": "OpenAI GPT models (current GPT-6 family in Codex)", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "OpenAI Codex is OpenAI's coding agent, available as an open-source CLI, an IDE extension and a cloud agent that works on tasks in parallel sandboxes. It is included with ChatGPT plans — from limited use on Free to Plus ($20/month) and Pro ($100–$500/month).", "about": [ "Codex can run locally (CLI or IDE extension, editing files on your machine) or in the cloud, where each task gets its own sandboxed environment and returns a diff or pull request. Cloud features include automatic GitHub code review and a Slack integration.", "Since April 2026 Codex usage is metered in credits aligned to token usage; plans include an allowance and you can buy more credits. You can also run the CLI with a plain OpenAI API key and pay per token." ], "strengths": [ "Included in ChatGPT subscriptions many developers already have", "Parallel cloud tasks that return PRs", "Open-source CLI", "Strong agentic benchmark results with current OpenAI models" ], "weaknesses": [ "OpenAI models only", "Credit-based usage rules are complex and change often", "Cloud features unavailable with API-key-only use" ], "best_for": "ChatGPT subscribers who want to delegate well-scoped tasks to background agents and review PRs.", "alternatives": [ "claude-code", "jules", "devin", "copilot" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=chatgpt.com&sz=128", "yc": false, "benchmarks": [ { "name": "Terminal-Bench 4.0", "score": "58.2%", "setup": "with GPT-6 Astra (max effort)", "date": "Oct 2026", "source": "https://www.tbench.ai/leaderboard" }, { "name": "Terminal-Bench 2.1", "score": "89.5%", "setup": "agent-level, Artificial Analysis harness", "date": "Aug 2026", "source": "https://www.morphllm.com/ai-coding-agent" } ], "recommendable": true }, { "slug": "opencode", "name": "OpenCode", "vendor": "OpenCode (open-source project)", "url": "https://opencode.ai/", "pricing_url": "https://opencode.ai/", "domain": "opencode.ai", "status": "active", "types": [ "cli", "extension" ], "primary": "cli", "interfaces": [ "Terminal (TUI)", "Desktop app", "IDE extension" ], "open_source": true, "license": "Open source (see GitHub repo)", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "OpenCode", "price": "Free", "note": "Bring your own model/provider; some free models included" }, { "name": "Zen / Go", "price": "Optional paid model access", "note": "Check official site" } ], "models": [ "claude", "openai", "gemini", "open", "local" ], "models_text": "75+ providers via Models.dev, local models, plus login with GitHub Copilot or ChatGPT Plus/Pro", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": null, "mcp": true, "self_host": true }, "summary": "OpenCode is a free, open-source AI coding agent for the terminal, desktop and IDE that works with 75+ model providers, local models, or your existing GitHub Copilot / ChatGPT subscription.", "about": [ "OpenCode offers a polished terminal UI with LSP integration, multiple parallel sessions on one project and shareable session links. It is model-agnostic: pick any provider per session, run local models, or sign in with Copilot or ChatGPT to use those subscriptions.", "The project reports over 200,000 GitHub stars and states that it does not store your code or context data. Optional paid offerings (Zen, Go) provide curated hosted model access." ], "strengths": [ "Free and open source with a large community", "Any model, including local — no vendor lock-in", "Can reuse Copilot or ChatGPT subscriptions", "Privacy-first: does not store your code" ], "weaknesses": [ "You manage model choice and costs yourself", "No built-in cloud/background agents", "Quality depends heavily on the model you pick" ], "best_for": "Developers who want a Claude-Code-style terminal agent without vendor lock-in.", "alternatives": [ "claude-code", "aider", "cline", "kilo-code" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=opencode.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "windsurf", "name": "Windsurf (now Devin Desktop)", "vendor": "Cognition", "url": "https://devin.ai/desktop", "pricing_url": "https://devin.ai/pricing/", "domain": "devin.ai", "status": "changed", "status_note": "On June 2, 2026 Cognition renamed Windsurf to Devin Desktop via an over-the-air update. Plans and pricing carried over; Windsurf's Cascade agent was replaced by Devin Local and reached end-of-life on July 1, 2026.", "types": [ "ide", "cloud" ], "primary": "ide", "interfaces": [ "Desktop editor (VS Code-compatible)", "CLI (Devin CLI)" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Free", "price": "$0", "note": "Light agent quota, unlimited Tab completions" }, { "name": "Pro", "price": "$20/mo", "note": "Frontier models, Devin Cloud access" }, { "name": "Max", "price": "$200/mo", "note": "Significantly higher quotas" }, { "name": "Teams", "price": "$80/mo + $40/mo per full seat", "note": "Up to 200 users" }, { "name": "Enterprise", "price": "Contact sales", "note": "SSO, VPC deployment" } ], "models": [ "claude", "openai", "gemini", "grok", "open" ], "models_text": "Frontier models from OpenAI, Anthropic, Google and SpaceXAI plus open-source models and Cognition's SWE models", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "Windsurf, the AI-native IDE, is now called Devin Desktop. Cognition rebranded it on June 2, 2026; it is the same VS Code-compatible editor with an Agent Command Center for local and cloud agents. Pricing: Free, Pro $20/month, Max $200/month, Teams $80/month + $40 per seat.", "about": [ "Devin Desktop keeps the Windsurf editor, extensions and keybindings but makes an Agent Command Center (a Kanban view of local and cloud agents) the default surface. Devin Local, a Rust rewrite, replaced Cascade as the local agent.", "It supports the Agent Client Protocol (ACP), so other agents such as Codex or Claude's agent can run inside the editor. Existing Windsurf subscriptions carried over unchanged." ], "strengths": [ "Full IDE plus a manager for many local and cloud agents", "Same subscription covers Devin Cloud on Pro and up", "Unlimited Tab completions even on Free", "ACP support for third-party agents" ], "weaknesses": [ "Brand and product churn (Windsurf → Devin Desktop; Cascade retired)", "Agent-manager-first UI is a change for editor-first users", "Quota-based limits" ], "best_for": "Former Windsurf users and teams who want one tool to run and supervise many agents.", "alternatives": [ "cursor", "devin", "kiro", "antigravity" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=devin.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "copilot", "name": "GitHub Copilot", "vendor": "GitHub (Microsoft)", "url": "https://github.com/features/copilot", "pricing_url": "https://github.com/features/copilot/plans", "domain": "github.com", "status": "active", "types": [ "extension", "cli", "cloud", "review" ], "primary": "extension", "interfaces": [ "VS Code, Visual Studio, JetBrains, Xcode, Eclipse, Neovim, Zed", "Copilot CLI", "github.com & GitHub Mobile" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 10, "tiers": [ { "name": "Free", "price": "$0", "note": "2,000 completions and 50 chat requests per month" }, { "name": "Pro", "price": "$10/mo", "note": "Unlimited completions; monthly AI credits" }, { "name": "Pro+", "price": "$39/mo", "note": "More AI credits" }, { "name": "Max", "price": "$100/mo", "note": "For sustained agent-driven work" }, { "name": "Business / Enterprise", "price": "Per seat", "note": "Check official site" } ], "models": [ "claude", "openai", "gemini", "grok", "open" ], "models_text": "Anthropic Claude, OpenAI GPT, Google Gemini, xAI Grok, Microsoft and Kimi models (varies by plan and feature)", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "GitHub Copilot is GitHub's AI coding assistant: inline completions, chat and agent mode in most major IDEs, a Copilot CLI, and a cloud agent you can assign issues to. Free tier available; Pro is $10/month, Pro+ $39/month and Max $100/month.", "about": [ "Copilot works as an extension in VS Code, Visual Studio, JetBrains IDEs, Xcode, Eclipse and more, and natively on github.com. You can assign an issue to the Copilot cloud agent and it opens a pull request, and Copilot can review PRs.", "Usage beyond completions is metered in GitHub AI Credits (1 credit = $0.01); completions and next-edit suggestions stay unlimited on paid plans. The model picker includes models from Anthropic, OpenAI, Google and xAI." ], "strengths": [ "Cheapest mainstream paid plan ($10/mo) with a usable free tier", "Works in nearly every IDE, not just VS Code", "Deep GitHub integration: issues → PRs, code review", "Wide multi-vendor model choice" ], "weaknesses": [ "Individual-plan interactions may be used for training unless you opt out", "Credit system makes heavy agent use harder to predict", "Agent quality depends on chosen model" ], "best_for": "Teams already on GitHub who want AI across editors and the PR workflow at a low price.", "alternatives": [ "cursor", "codex", "jetbrains-ai", "tabnine" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=github.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "aider", "name": "Aider", "vendor": "Paul Gauthier & contributors (open source)", "url": "https://aider.chat/", "pricing_url": "https://aider.chat/", "domain": "aider.chat", "status": "active", "types": [ "cli" ], "primary": "cli", "interfaces": [ "Terminal", "Works with any editor via file-watch comments" ], "open_source": true, "license": "Apache-2.0", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "Aider", "price": "Free", "note": "Pay only your model provider (BYOK)" } ], "models": [ "claude", "openai", "gemini", "open", "local" ], "models_text": "Almost any LLM via API keys, including local models", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": false, "mcp": null, "self_host": true }, "summary": "Aider is a free, open-source AI pair-programming tool for the terminal. It edits files in your local git repo, auto-commits each change with a sensible message, and works with almost any LLM through your own API keys.", "about": [ "Aider builds a map of your whole repository so the model has context in larger projects, then applies edits and commits them to git so every AI change is easy to diff and undo. It supports 100+ languages, images and web pages as context, and voice input.", "It is one of the longest-running open-source coding agents, with tens of thousands of GitHub stars. You pay only for model usage." ], "strengths": [ "Git-native: every change is a commit you can review or revert", "Model-agnostic, including local models", "Lightweight and scriptable", "Free and open source" ], "weaknesses": [ "More pair-programmer than autonomous agent", "Terminal-only UI", "You manage API keys and costs" ], "best_for": "Git-centric developers who want precise, reviewable AI edits with any model.", "alternatives": [ "opencode", "claude-code", "cline" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=aider.chat&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "antigravity", "name": "Google Antigravity", "vendor": "Google", "url": "https://antigravity.google/", "pricing_url": "https://antigravity.google/pricing", "domain": "antigravity.google", "status": "active", "status_note": "Gemini CLI stopped serving free, Google AI Pro and Ultra individual users on June 18, 2026; Google moved those users to Antigravity CLI.", "types": [ "ide", "cli" ], "primary": "ide", "interfaces": [ "Antigravity desktop app / IDE", "Antigravity CLI", "IDE extensions", "SDK" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": null, "tiers": [ { "name": "Individual", "price": "$0", "note": "Unlimited Tab completions and Command requests; basic weekly agent limits" }, { "name": "Google AI Pro", "price": "Via Google AI Pro subscription", "note": "Higher limits — check official site for price" }, { "name": "Google AI Ultra", "price": "Via Google AI Ultra subscription", "note": "Highest limits" }, { "name": "Business (Google Cloud)", "price": "Gemini Enterprise from $30/seat/mo, or pay-as-you-go", "note": "Consumption-based via API also available" } ], "models": [ "gemini" ], "models_text": "Google Gemini models (plus a selection of other agent models — check site)", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": null, "pr_review": null, "mcp": true, "self_host": false }, "summary": "Google Antigravity is Google's agent-first development platform — a desktop IDE, CLI and extensions built around Gemini models. The Individual tier is free with weekly agent limits; higher limits come with Google AI Pro or Ultra subscriptions.", "about": [ "Antigravity lets agents plan and execute work across the editor, terminal and browser while you review their artifacts. The Antigravity CLI is included on every tier and shares the same quota pool.", "For organizations, Antigravity is available through Google Cloud: Gemini Enterprise plans start at $30 per seat per month, or pay-as-you-go with consumption pricing." ], "strengths": [ "Generous free individual tier", "Tight integration with Gemini models and Google Cloud", "IDE and CLI share one quota", "Enterprise path via Google Cloud terms" ], "weaknesses": [ "Mostly tied to Google's model ecosystem", "Product naming has shifted (Gemini CLI → Antigravity CLI)", "Paid limits depend on Google AI subscription tiers" ], "best_for": "Developers in the Google/Gemini ecosystem who want a free agent-first IDE.", "alternatives": [ "cursor", "windsurf", "kiro", "jules" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=antigravity.google&sz=128", "yc": false, "benchmarks": [ { "name": "Terminal-Bench 2.1", "score": "70.7%", "setup": "agent-level, listed as 'Gemini CLI / Antigravity', Artificial Analysis harness", "date": "Aug 2026", "source": "https://www.morphllm.com/ai-coding-agent" } ], "recommendable": true }, { "slug": "warp", "name": "Warp", "vendor": "Warp", "url": "https://www.warp.dev/", "pricing_url": "https://www.warp.dev/pricing", "domain": "warp.dev", "status": "active", "types": [ "cli", "cloud" ], "primary": "cli", "interfaces": [ "Terminal app (macOS, Linux, Windows)", "Cloud agents" ], "open_source": true, "license": "Client source published on GitHub (AGPL-3.0 / MIT)", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Free", "price": "$0", "note": "Terminal plus limited AI" }, { "name": "Build", "price": "$20/mo", "note": "$20 of agent usage included at API rates" }, { "name": "Max", "price": "$200/mo", "note": "$200 of agent usage included" }, { "name": "Business", "price": "$50/user/mo", "note": "Up to 25 seats, SAML SSO" }, { "name": "Enterprise", "price": "Custom", "note": "Self-hosted cloud agents, admin controls" } ], "models": [ "claude", "openai", "open" ], "models_text": "Frontier and open-source models from OpenAI, Anthropic, z.ai and others", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": null, "mcp": true, "self_host": false }, "summary": "Warp is a modern terminal with a built-in coding agent: natural-language commands, multi-step agent tasks and cloud agents, using models from OpenAI, Anthropic and others. Free tier available; Build is $20/month with $20 of agent usage included.", "about": [ "Warp rethinks the terminal with blocks, an editor-like input and an agent that can run commands, edit code and debug from the same window. Paid plans include agent usage billed at API rates, so you see exactly what you consume.", "Business and Enterprise plans add SSO, admin controls, and on Enterprise, self-hosted cloud agents on your own infrastructure." ], "strengths": [ "Great terminal UX even without AI", "Transparent usage-at-API-rates pricing", "Agent works where commands actually run", "Multi-model" ], "weaknesses": [ "Requires switching terminal apps", "Account sign-in for AI features", "Heavier than a plain terminal" ], "best_for": "Terminal-heavy developers who want AI help for both shell work and code changes.", "alternatives": [ "claude-code", "opencode", "codex" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=warp.dev&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "openhands", "name": "OpenHands", "vendor": "All Hands AI", "url": "https://openhands.dev/", "pricing_url": "https://openhands.dev/pricing", "domain": "openhands.dev", "status": "active", "types": [ "cli", "cloud" ], "primary": "cloud", "interfaces": [ "Web GUI (local or cloud)", "CLI", "SDK" ], "open_source": true, "license": "Open source (MIT core)", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "Open source (local)", "price": "Free", "note": "Run OpenHands yourself" }, { "name": "Cloud Individual", "price": "Free", "note": "BYOK or providers at cost" }, { "name": "Enterprise", "price": "Custom", "note": "SaaS or self-hosted in your VPC, SSO" } ], "models": [ "claude", "openai", "gemini", "open", "local" ], "models_text": "Model-agnostic: bring your own key or use providers at cost", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": null, "mcp": true, "self_host": true }, "summary": "OpenHands (formerly OpenDevin) is an open-source autonomous coding agent platform from All Hands AI. Run it free locally or in their cloud with your own model keys; Enterprise adds self-hosting in your VPC and SSO.", "about": [ "OpenHands agents can write code, run commands and browse the web inside a sandbox, and you can drive them from a web GUI, a CLI or an SDK for building your own agent workflows.", "Because it is model-agnostic and self-hostable, it is a common choice for teams that need autonomy without sending code to a closed vendor." ], "strengths": [ "Open source and self-hostable", "Model-agnostic", "Free cloud tier with BYOK", "SDK for custom agent workflows" ], "weaknesses": [ "More setup than commercial tools", "Results vary with the model you bring", "Less polished IDE integration" ], "best_for": "Teams that want a Devin-style autonomous agent they can self-host.", "alternatives": [ "devin", "opencode", "jules" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=openhands.dev&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "zed", "name": "Zed", "vendor": "Zed Industries", "url": "https://zed.dev/", "pricing_url": "https://zed.dev/pricing", "domain": "zed.dev", "status": "active", "types": [ "ide" ], "primary": "ide", "interfaces": [ "Native desktop editor (macOS, Linux, Windows)" ], "open_source": true, "license": "Open source (GPL/AGPL/Apache components)", "pricing": "freemium", "free_tier": true, "price_from": 10, "tiers": [ { "name": "Personal", "price": "$0", "note": "2,000 accepted edit predictions; unlimited with your own keys or external agents" }, { "name": "Pro", "price": "$10/mo", "note": "Unlimited edit predictions, $5 of tokens, then API list price +10%" }, { "name": "Business", "price": "$30/seat/mo", "note": "Org AI policies, data governance" } ], "models": [ "claude", "openai", "gemini", "open", "local" ], "models_text": "Zed-hosted models, or your own keys (Anthropic, OpenAI, Google, Bedrock, Ollama, OpenRouter and more); external agents via ACP", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": true, "self_host": false }, "summary": "Zed is a fast, open-source code editor written in Rust with built-in AI: edit predictions, an agent panel and support for external agents like Claude's agent and Codex CLI. Free with your own API keys; Pro is $10/month.", "about": [ "Zed focuses on raw performance and real-time multiplayer collaboration, with AI as an optional layer you can disable entirely. It pioneered the Agent Client Protocol so third-party agents can run inside the editor.", "Pro bills hosted model use at API list price plus 10% after the included $5; Business adds org-wide AI and data-sharing controls." ], "strengths": [ "Extremely fast native editor", "Open source; AI can be fully disabled", "Bring any key or external agent", "Cheap Pro plan" ], "weaknesses": [ "Smaller extension ecosystem than VS Code", "SSO/SCIM not yet available on Business", "Agent features less mature than Cursor" ], "best_for": "Developers who want a fast, lean editor and control over which AI they use.", "alternatives": [ "cursor", "windsurf", "cline" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=zed.dev&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "pearai", "name": "PearAI", "vendor": "PearAI (YC)", "url": "https://trypear.ai/", "pricing_url": "https://trypear.ai/", "domain": "trypear.ai", "status": "sunset", "status_note": "PearAI appears inactive: its main repository was archived in June 2026 and PitchBook lists the company as out of business. We found no official shutdown notice — verify before relying on it.", "types": [ "ide" ], "primary": "ide", "interfaces": [ "Desktop editor (VS Code fork)" ], "open_source": true, "license": "Open source (VS Code fork)", "pricing": "unknown", "free_tier": null, "price_from": null, "tiers": [ { "name": "Pricing", "price": "Check official site", "note": "Product appears inactive" } ], "models": [ "claude", "openai" ], "models_text": "Multiple hosted models (historically)", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": null, "self_host": false }, "summary": "PearAI was a YC-backed, open-source AI code editor forked from VS Code. As of October 2026 it appears discontinued (repository archived June 2026); consider Cursor, Zed or Cline instead.", "about": [ "PearAI positioned itself as an open-source alternative to Cursor, bundling AI chat, autocomplete and agent tools into a VS Code fork.", "With development apparently stopped, existing builds may still run but should not be expected to receive fixes or security patches." ], "strengths": [ "Open-source codebase", "Familiar VS Code base" ], "weaknesses": [ "Appears no longer maintained", "No security updates expected" ], "best_for": "Nobody starting fresh — pick an actively maintained editor.", "alternatives": [ "cursor", "zed", "cline" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=trypear.ai&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "supermaven", "name": "Supermaven", "vendor": "Supermaven (acquired by Cursor)", "url": "https://supermaven.com/", "pricing_url": "https://supermaven.com/blog/sunsetting-supermaven", "domain": "supermaven.com", "status": "sunset", "status_note": "Cursor acquired Supermaven in November 2024 and sunset the product on November 21, 2025. Existing Neovim and JetBrains customers keep free autocomplete; VS Code users were told to migrate to Cursor.", "types": [ "extension" ], "primary": "extension", "interfaces": [ "VS Code, JetBrains, Neovim plugins (legacy)" ], "open_source": false, "license": "Proprietary", "pricing": "free", "free_tier": true, "price_from": null, "tiers": [ { "name": "Existing customers", "price": "Free autocomplete", "note": "No new paid plans; refunds issued at sunset" } ], "models": [ "proprietary" ], "models_text": "Supermaven's own long-context completion model", "byok": false, "features": { "agent": false, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": false, "self_host": false }, "summary": "Supermaven was a fast AI autocomplete plugin known for its long context window. Cursor acquired it in 2024 and sunset it on November 21, 2025; its technology now powers Cursor Tab.", "about": [ "Supermaven launched in February 2024 and won fans for very low-latency completions that understood large codebases.", "After the sunset, agent chat was discontinued and existing customers were refunded; free autocomplete continues for existing Neovim and JetBrains users." ], "strengths": [ "Very fast completions (legacy users)" ], "weaknesses": [ "Discontinued for new users", "No agent features" ], "best_for": "Existing Neovim/JetBrains users only. New users: Cursor Tab or Copilot.", "alternatives": [ "cursor", "copilot", "tabnine" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=supermaven.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "cline", "name": "Cline", "vendor": "Cline Bot Inc.", "url": "https://cline.bot/", "pricing_url": "https://cline.bot/pricing", "domain": "cline.bot", "status": "active", "types": [ "extension", "cli" ], "primary": "extension", "interfaces": [ "VS Code extension", "CLI", "JetBrains (Enterprise)", "SDK" ], "open_source": true, "license": "Apache-2.0", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "Open source", "price": "Free", "note": "Pay only for inference (BYOK or Cline provider at cost)" }, { "name": "Enterprise", "price": "Custom", "note": "SSO, SLA, team dashboard, JetBrains extension" } ], "models": [ "claude", "openai", "gemini", "grok", "open", "local" ], "models_text": "Any provider via your own API keys, plus local models", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": false, "mcp": true, "self_host": true }, "summary": "Cline is a free, open-source autonomous coding agent for VS Code (and a CLI) that plans, edits files and runs commands with your approval. You pay only for model usage via your own API keys.", "about": [ "Cline popularised the plan/act workflow inside VS Code: it proposes a plan, then executes step by step with diffs and terminal commands you approve. It has an MCP marketplace for adding tools.", "Its client-side architecture means code goes straight from your machine to your chosen model provider. Enterprise adds SSO, centralised billing and role-based controls." ], "strengths": [ "Free and open source", "Any model, no markup when using your keys", "Transparent step-by-step approvals", "Large MCP ecosystem" ], "weaknesses": [ "Token costs can climb on big tasks", "No built-in autocomplete", "JetBrains support is Enterprise-only" ], "best_for": "VS Code users who want an autonomous agent with full control over models and spend.", "alternatives": [ "kilo-code", "opencode", "copilot", "continue" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=cline.bot&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "kilo-code", "name": "Kilo Code", "vendor": "Kilo Code", "url": "https://kilocode.ai/", "pricing_url": "https://kilocode.ai/pricing", "domain": "kilocode.ai", "status": "active", "types": [ "extension", "cli" ], "primary": "extension", "interfaces": [ "VS Code extension", "JetBrains extension", "CLI", "Kilo Desktop" ], "open_source": true, "license": "Open source", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "Free & open source", "price": "$0", "note": "Usage billed separately (BYOK, local or Kilo Gateway)" }, { "name": "Teams", "price": "$15/user/mo", "note": "Team management, shared BYOK, analytics" }, { "name": "Enterprise", "price": "Custom", "note": "SSO, audit logs, SLA" }, { "name": "Kilo Pass", "price": "From $19/mo (optional)", "note": "Prepaid inference" } ], "models": [ "claude", "openai", "gemini", "grok", "open", "local" ], "models_text": "Any AI model: BYOK, local models, or the pay-as-you-go Kilo Gateway", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": null, "pr_review": null, "mcp": true, "self_host": true }, "summary": "Kilo Code is an open-source AI coding agent for VS Code, JetBrains and the CLI, with modes for architecting, coding, debugging and asking. The extension is free; Teams costs $15/user/month and model usage is billed separately.", "about": [ "Kilo Code descends from the Cline/Roo Code family of open-source agents and adds autocomplete, shared agent modes for teams, and its own Kilo Gateway for pay-as-you-go inference.", "You can bring your own keys, run local models, or buy an optional Kilo Pass." ], "strengths": [ "Open source across VS Code, JetBrains and CLI", "Works with any model", "Cheap team plan", "Specialised modes (architect/code/debug)" ], "weaknesses": [ "Many overlapping options (Gateway, Pass, BYOK) to understand", "You manage model costs" ], "best_for": "Teams wanting an open-source agent across VS Code and JetBrains with light admin features.", "alternatives": [ "cline", "opencode", "continue" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=kilocode.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "continue", "name": "Continue", "vendor": "Continue (acquired by Cursor)", "url": "https://continue.dev/", "pricing_url": "https://continue.dev/", "domain": "continue.dev", "status": "sunset", "status_note": "Per multiple reports, Cursor acquired Continue in June 2026 and the hosted Continue Hub shut down on July 15, 2026. The Apache-2.0 extension can still be installed and used with local models or your own keys, but is reported to be no longer actively maintained.", "types": [ "extension" ], "primary": "extension", "interfaces": [ "VS Code extension", "JetBrains extension", "CLI" ], "open_source": true, "license": "Apache-2.0", "pricing": "free", "free_tier": true, "price_from": 0, "tiers": [ { "name": "Open-source extension", "price": "Free", "note": "Use with local models or your own API keys" }, { "name": "Hosted plans", "price": "Discontinued", "note": "Hub shut down July 15, 2026" } ], "models": [ "claude", "openai", "gemini", "open", "local" ], "models_text": "Any model via your own keys, including Ollama/local", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": true, "self_host": true }, "summary": "Continue is an open-source (Apache-2.0) AI assistant for VS Code and JetBrains. After Cursor's reported acquisition in June 2026 its hosted Hub closed on July 15, 2026; the extension still works with local models or your own keys.", "about": [ "Continue offered chat, autocomplete and agent mode configurable with any model, which made it popular with teams running local models via Ollama.", "If you need active maintenance, consider Cline, Kilo Code or Copilot; if you need fully local AI, Continue's extension remains usable." ], "strengths": [ "Open source; works fully local", "JetBrains and VS Code support" ], "weaknesses": [ "Hosted service discontinued", "Reportedly no longer actively maintained" ], "best_for": "Existing users with local-model setups.", "alternatives": [ "cline", "kilo-code", "tabnine" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=continue.dev&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "devin", "name": "Devin", "vendor": "Cognition", "url": "https://devin.ai/", "pricing_url": "https://devin.ai/pricing/", "domain": "devin.ai", "status": "active", "types": [ "cloud", "ide", "cli", "review" ], "primary": "cloud", "interfaces": [ "Devin Cloud (web)", "Devin Desktop", "Devin CLI", "Slack / Teams, Linear / Jira, GitHub / GitLab / Bitbucket" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Free", "price": "$0", "note": "Light agent quota (Desktop)" }, { "name": "Pro", "price": "$20/mo", "note": "Frontier models, Devin Cloud access" }, { "name": "Max", "price": "$200/mo", "note": "Much higher quotas" }, { "name": "Teams", "price": "$80/mo + $40/mo per full seat", "note": "Up to 200 users, unlimited concurrent sessions" }, { "name": "Enterprise", "price": "Contact sales", "note": "VPC deployment, SSO" } ], "models": [ "claude", "openai", "gemini", "grok", "open" ], "models_text": "Cognition's SWE models plus frontier models from OpenAI, Anthropic, Google and SpaceXAI", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "Devin is Cognition's autonomous AI software engineer. Assign it a task from Slack, Jira, Linear or GitHub and it plans, codes and tests in its own cloud machine, then opens a pull request. Plans start free; Pro is $20/month, Teams $80/month + $40 per seat.", "about": [ "Devin now spans Devin Cloud (long-running autonomous agent), Devin Desktop (the former Windsurf IDE), Devin CLI and Devin Review for code review on every diff. One subscription covers the family.", "DeepWiki and Ask Devin help it (and you) understand large codebases before it starts changing them." ], "strengths": [ "Truly asynchronous: hand off tasks and get PRs", "Integrates with Slack, Jira, Linear and all major Git hosts", "Same plan covers IDE, CLI and cloud", "VPC deployment for enterprises" ], "weaknesses": [ "Autonomous output still needs careful review", "Quota-based limits", "Closed source" ], "best_for": "Teams with a backlog of well-defined tickets they want to delegate.", "alternatives": [ "codex", "jules", "factory", "openhands" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=devin.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "replit", "name": "Replit Agent", "vendor": "Replit (YC)", "url": "https://replit.com/", "pricing_url": "https://replit.com/pricing", "domain": "replit.com", "status": "active", "types": [ "builder", "cloud" ], "primary": "builder", "interfaces": [ "Browser IDE", "Mobile app" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Starter", "price": "Free", "note": "Try Replit Agent with limits — check site" }, { "name": "Core", "price": "$20/mo ($18/mo billed annually)", "note": "$20 of monthly usage credits, up to 5 collaborators" }, { "name": "Pro / Teams / Enterprise", "price": "Higher tiers", "note": "Check official site" } ], "models": [ "proprietary" ], "models_text": "Curated hosted models chosen by Replit — check official site", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "Replit Agent builds, runs and deploys full apps from a prompt inside Replit's browser IDE — no local setup. Core costs $20/month ($18 billed annually) and includes $20 of usage credits.", "about": [ "Replit bundles editor, hosting, databases and deployment, so the Agent can go from idea to a live URL in one place. It suits prototypes, internal tools and learning.", "Usage beyond included credits is billed by consumption; check Replit's pricing page for current Pro and Teams tiers." ], "strengths": [ "Zero setup: everything runs in the browser", "Hosting and deployment built in", "Good for non-experts and prototypes" ], "weaknesses": [ "Less control than a local dev setup", "Usage-based costs can grow with agent work", "Tied to Replit's hosting" ], "best_for": "Builders who want a working, hosted app fast without configuring a machine.", "alternatives": [ "lovable", "bolt", "v0" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=replit.com&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "lovable", "name": "Lovable", "vendor": "Lovable", "url": "https://lovable.dev/", "pricing_url": "https://lovable.dev/pricing", "domain": "lovable.dev", "status": "active", "types": [ "builder" ], "primary": "builder", "interfaces": [ "Web app" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 25, "tiers": [ { "name": "Free", "price": "$0", "note": "5 build credits/day (up to 30/month) + 20 Cloud credits" }, { "name": "Pro", "price": "From $25/mo (100 credits)", "note": "Scales by credit tier; $21/mo billed annually" }, { "name": "Business", "price": "From $50/mo", "note": "Team features; scales by credit tier" }, { "name": "Enterprise", "price": "Custom", "note": "Volume credit pricing" } ], "models": [ "proprietary" ], "models_text": "Hosted models chosen by Lovable", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "Lovable is an AI app builder: describe a web app in plain English and it generates a working full-stack app you can edit by chat, sync to GitHub and host on Lovable Cloud. Free tier available; Pro starts at $25/month for 100 credits.", "about": [ "Lovable targets founders, designers and product people who want a real, editable codebase rather than a mockup. Credits pay for building messages, Cloud hosting and AI features inside your app.", "Workspaces have unlimited members on every plan; you pay for credits, not seats." ], "strengths": [ "Very fast idea-to-app", "Real code you can export to GitHub", "No per-seat pricing" ], "weaknesses": [ "Credit costs scale with complex apps", "Less suited to large existing codebases", "Credits expire" ], "best_for": "Non-engineers and founders prototyping full-stack web apps.", "alternatives": [ "bolt", "v0", "replit" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=lovable.dev&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "v0", "name": "v0", "vendor": "Vercel", "url": "https://v0.app/", "pricing_url": "https://v0.app/pricing", "domain": "v0.app", "status": "active", "types": [ "builder" ], "primary": "builder", "interfaces": [ "Web app", "iOS app" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 30, "tiers": [ { "name": "Free", "price": "$0", "note": "7 messages/day; deploy to Vercel" }, { "name": "Plus", "price": "$30/user/mo", "note": "$30 monthly credits + $2 daily login credits" }, { "name": "Business", "price": "$100/user/mo", "note": "Training opt-out by default" }, { "name": "Enterprise", "price": "Custom", "note": "SAML SSO, RBAC, SLAs" } ], "models": [ "proprietary" ], "models_text": "v0's own model family (Mini / Pro / Max) running on Vercel", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "v0 by Vercel generates React/Next.js UIs and full-stack apps from prompts, with visual Design Mode, GitHub sync and one-click deploys to Vercel. Free tier with 7 messages/day; Plus is $30/user/month.", "about": [ "v0 is strongest at front-end: polished components and pages using modern React conventions that drop straight into a Next.js project.", "Paid plans include monthly credits; additional credits can be bought and shared across a team." ], "strengths": [ "High-quality React/Next.js UI output", "Seamless Vercel deployment", "GitHub sync and visual editing" ], "weaknesses": [ "Best fit is the Vercel/Next.js stack", "Per-user pricing adds up for teams" ], "best_for": "Front-end developers and designers building on Next.js and Vercel.", "alternatives": [ "lovable", "bolt", "replit" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=v0.app&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "bolt", "name": "Bolt", "vendor": "StackBlitz", "url": "https://bolt.new/", "pricing_url": "https://bolt.new/pricing", "domain": "bolt.new", "status": "active", "types": [ "builder" ], "primary": "builder", "interfaces": [ "Web app (in-browser WebContainers)" ], "open_source": false, "license": "Proprietary (bolt.diy is the OSS sibling)", "pricing": "freemium", "free_tier": true, "price_from": 25, "tiers": [ { "name": "Free", "price": "$0", "note": "300K tokens/day, 1M tokens/month" }, { "name": "Pro", "price": "$25/mo", "note": "From 10M tokens/month, no daily limit, custom domains" }, { "name": "Teams", "price": "$30/member/mo", "note": "Admin and design-system features" }, { "name": "Enterprise", "price": "Custom", "note": "SSO, SLAs" } ], "models": [ "proprietary" ], "models_text": "Hosted models chosen by Bolt — check official site", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "Bolt (bolt.new) by StackBlitz builds and runs full-stack web apps from a prompt entirely in your browser, with hosting, databases and custom domains. Free tier with 1M tokens/month; Pro is $25/month.", "about": [ "Bolt runs a real Node.js environment in the browser via StackBlitz WebContainers, so the generated app installs packages and runs instantly.", "Unused Pro tokens roll over to the next month." ], "strengths": [ "Instant in-browser dev environment", "Hosting and databases included", "Token rollover on Pro" ], "weaknesses": [ "Token-based pricing is hard to predict", "Better for new apps than big existing repos" ], "best_for": "Quick full-stack prototypes and MVPs without local setup.", "alternatives": [ "lovable", "v0", "replit" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=bolt.new&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "pythagora", "name": "Pythagora (now Pazi)", "vendor": "Pythagora Technologies (YC)", "url": "https://pazi.ai/", "pricing_url": "https://pazi.ai/", "domain": "pazi.ai", "status": "changed", "status_note": "In 2026 Pythagora pivoted and rebranded its product as Pazi, a platform for autonomous AI-run businesses and workflow automation. It is no longer primarily an AI coding tool.", "types": [ "builder" ], "primary": "builder", "interfaces": [ "Web app" ], "open_source": false, "license": "Proprietary (the earlier GPT Pilot project was open source)", "pricing": "freemium", "free_tier": true, "price_from": null, "tiers": [ { "name": "Pazi plans", "price": "Free and paid 'funding' tiers", "note": "Check official site" } ], "models": [ "proprietary" ], "models_text": "Hosted models", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "Pythagora was a YC-backed AI app builder (and the team behind the open-source GPT Pilot). In 2026 it pivoted to Pazi, a platform for building autonomous AI-run businesses, so it is no longer a coding-agent pick.", "about": [ "Pythagora let you build full-stack apps by chatting with a team of agents inside VS Code or the web.", "After the pivot, people looking for an AI app builder should compare Lovable, Bolt, v0 or Replit instead." ], "strengths": [ "Pioneered multi-agent app building" ], "weaknesses": [ "No longer focused on coding", "Plans changed with the pivot" ], "best_for": "Not recommended as a coding tool after the 2026 pivot.", "alternatives": [ "lovable", "bolt", "replit" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=pazi.ai&sz=128", "yc": true, "benchmarks": [], "recommendable": false }, { "slug": "coderabbit", "name": "CodeRabbit", "vendor": "CodeRabbit", "url": "https://www.coderabbit.ai/", "pricing_url": "https://www.coderabbit.ai/pricing", "domain": "coderabbit.ai", "status": "active", "types": [ "review" ], "primary": "review", "interfaces": [ "GitHub, GitLab and other Git hosts (PR comments)", "CLI", "IDE" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 24, "tiers": [ { "name": "Essentials", "price": "$24/user/mo", "note": "Agentic PR, CLI and IDE reviews" }, { "name": "Team", "price": "$48/user/mo", "note": "Adds test generation, PR queue prioritisation" }, { "name": "Advanced", "price": "$72/user/mo", "note": "Most advanced analysis and monitoring" }, { "name": "Enterprise", "price": "Contact sales", "note": "Enterprise controls" } ], "pricing_note": "Annual plans save 20%; 14-day free trial.", "models": [ "proprietary" ], "models_text": "Vendor-managed frontier models", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": null }, "summary": "CodeRabbit is an AI code review tool that posts line-by-line reviews, summaries and one-click fixes on every pull request, and also reviews in the CLI and IDE. Plans start at $24/user/month with a 14-day trial.", "about": [ "CodeRabbit learns from your feedback and team conventions, runs pre-merge checks and flags 'AI slop' in agent-written code. It can also explain and document each change.", "Higher tiers add automated test generation, post-merge actions and a prioritised PR queue." ], "strengths": [ "Reviews on PRs, CLI and IDE from one tool", "Learns team conventions over time", "Good fit for reviewing agent-generated PRs" ], "weaknesses": [ "Per-seat cost adds up", "Can be noisy until tuned" ], "best_for": "Teams that want an automated first-pass reviewer on every PR.", "alternatives": [ "greptile", "graphite", "qodo", "bito" ], "status_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=coderabbit.ai&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "greptile", "name": "Greptile", "vendor": "Greptile (YC)", "url": "https://www.greptile.com/", "pricing_url": "https://www.greptile.com/pricing", "domain": "greptile.com", "status": "active", "types": [ "review" ], "primary": "review", "interfaces": [ "GitHub, GitLab, Bitbucket, Gitea (PR comments)", "CLI" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 30, "tiers": [ { "name": "Starter", "price": "Free", "note": "50 credits/month, 1 active developer" }, { "name": "Pro", "price": "$30/seat/mo", "note": "50 credits/seat, extra credits $1 each" }, { "name": "Enterprise", "price": "Custom", "note": "Self-hosting, SSO/SAML, data residency" } ], "pricing_note": "50% startup discount available for early-stage startups.", "models": [ "proprietary" ], "models_text": "Vendor-managed models", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": true }, "summary": "Greptile is an AI code reviewer that indexes your whole codebase to catch bugs in pull requests with full-repo context. Free Starter plan; Pro is $30/seat/month; Enterprise can self-host.", "about": [ "Because Greptile builds a graph of your entire repository, its reviews can flag issues that span files and services rather than only the diff.", "It supports custom review rules and learns from your team's feedback." ], "strengths": [ "Whole-codebase context in reviews", "Self-hostable on Enterprise", "Free tier for solo devs" ], "weaknesses": [ "Credit system on top of seats", "Mostly review-only (not a coding agent)" ], "best_for": "Teams with large codebases where diff-only review misses cross-file bugs.", "alternatives": [ "coderabbit", "graphite", "qodo" ], "status_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=greptile.com&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "ellipsis", "name": "Ellipsis", "vendor": "Ellipsis (YC)", "url": "https://www.ellipsis.dev/", "pricing_url": "https://www.ellipsis.dev/pricing", "domain": "ellipsis.dev", "status": "active", "status_note": "Ellipsis has broadened from an AI code reviewer into a platform for cloud coding agents.", "types": [ "cloud", "review" ], "primary": "cloud", "interfaces": [ "GitHub", "Slack", "Your AWS account (VPC)" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": null, "tiers": [ { "name": "Managed SaaS", "price": "Usage-based", "note": "Tokens at cost + 10% platform fee; $0 per seat" }, { "name": "Your AWS account", "price": "Usage-based + AWS costs", "note": "Deploys into your VPC" } ], "models": [ "claude", "openai" ], "models_text": "Frontier models; interactive sessions can run on your Claude Code or Codex subscription", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": true, "mcp": null, "self_host": true }, "summary": "Ellipsis runs cloud coding agents that review PRs, fix bugs and answer questions from GitHub and Slack. There are no seat fees: you pay tokens at cost plus a 10% platform fee, and it can deploy inside your own AWS VPC.", "about": [ "Each agent session runs in its own sandbox. Agents are configured in YAML, and interactive sessions such as @ellipsis in Slack can use your existing Claude Code or Codex subscription.", "Teams can start on the managed SaaS and move to their own AWS account later with the same agents and configs." ], "strengths": [ "No per-seat fees", "Can run entirely in your AWS VPC", "Reuses existing Claude Code/Codex subscriptions" ], "weaknesses": [ "Usage-based cost is less predictable", "More setup than a plug-and-play reviewer" ], "best_for": "Engineering teams that want configurable review and fix agents under their own cloud controls.", "alternatives": [ "coderabbit", "devin", "codex" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=ellipsis.dev&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "graphite", "name": "Graphite", "vendor": "Graphite (acquiring company: Cursor)", "url": "https://graphite.com/", "pricing_url": "https://graphite.com/docs/billing-plans", "domain": "graphite.com", "status": "active", "status_note": "Cursor announced a definitive agreement to acquire Graphite on December 19, 2025; Graphite continues to operate as a product.", "types": [ "review", "cli" ], "primary": "review", "interfaces": [ "GitHub", "Graphite CLI", "Web dashboard", "VS Code extension" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Hobby", "price": "Free", "note": "Personal projects" }, { "name": "Starter", "price": "$20/seat/mo (billed annually)", "note": "Limited AI reviews" }, { "name": "Team", "price": "$40/seat/mo (billed annually)", "note": "Unlimited AI reviews and chat" }, { "name": "Enterprise", "price": "Custom", "note": "Check official site" } ], "models": [ "proprietary" ], "models_text": "Vendor-managed models", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": false }, "summary": "Graphite is a code review platform built around stacked pull requests, with an AI reviewer (Graphite Agent), a merge queue and a CLI. Free Hobby plan; Starter is $20/seat/month and Team $40/seat/month, billed annually.", "about": [ "Stacked PRs let you split big changes into small, dependent pull requests that are easier to review — a natural fit for the volume of code AI agents produce.", "Graphite was announced as being acquired by Cursor in December 2025." ], "strengths": [ "Stacked PR workflow + AI review in one", "Fast review UI and merge queue", "Free hobby tier" ], "weaknesses": [ "GitHub-only", "Annual billing for listed prices" ], "best_for": "High-velocity teams on GitHub that want smaller PRs and faster reviews.", "alternatives": [ "coderabbit", "greptile", "qodo" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=graphite.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "tabnine", "name": "Tabnine", "vendor": "Tabnine", "url": "https://www.tabnine.com/", "pricing_url": "https://www.tabnine.com/pricing/", "domain": "tabnine.com", "status": "active", "types": [ "extension", "cli" ], "primary": "extension", "interfaces": [ "VS Code, JetBrains, Visual Studio, Eclipse and more", "Tabnine CLI" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 39, "tiers": [ { "name": "Code Assistant", "price": "$39/user/mo (annual)", "note": "Completions and chat in the IDE" }, { "name": "Agentic Platform", "price": "$59/user/mo (annual)", "note": "Autonomous agents, Tabnine CLI, Context Engine" }, { "name": "Headless Agents", "price": "Capacity-based", "note": "Separate pricing" } ], "pricing_note": "Prices from tabnine.com as reported in search results; the page did not render for our fetcher — confirm on the official site.", "models": [ "claude", "openai", "gemini", "open", "proprietary" ], "models_text": "Choice of vendor and private models; supports self-hosted/private deployment", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": true }, "summary": "Tabnine is an enterprise-focused AI coding assistant known for privacy: completions, chat and agents in most IDEs, with SaaS, VPC or on-prem deployment. Listed plans start at $39/user/month (annual).", "about": [ "Tabnine's pitch is control: deploy where you want, choose which models are allowed, and keep code out of third-party training. Its Context Engine connects agents to your organisation's code and tools.", "In 2026 it split pricing into Code Assistant and Agentic Platform tiers and added a CLI agent." ], "strengths": [ "Air-gapped / on-prem deployment options", "Broad IDE coverage incl. Eclipse & Visual Studio", "Strong governance and model controls" ], "weaknesses": [ "Pricier than Copilot for individuals", "No free tier" ], "best_for": "Regulated enterprises that need private deployment.", "alternatives": [ "copilot", "jetbrains-ai", "augment" ], "status_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=tabnine.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "qodo", "name": "Qodo", "vendor": "Qodo (formerly CodiumAI)", "url": "https://www.qodo.ai/", "pricing_url": "https://www.qodo.ai/pricing/", "domain": "qodo.ai", "status": "active", "types": [ "review", "extension", "cli" ], "primary": "review", "interfaces": [ "Git (GitHub, GitLab, Bitbucket, Gerrit on Enterprise)", "IDE plugins", "CLI" ], "open_source": false, "license": "Proprietary (PR-Agent is open source)", "pricing": "paid", "free_tier": false, "price_from": 30, "tiers": [ { "name": "Pro Team", "price": "$30/mo (credit packs, up to 30 users)", "note": "14-day free trial, no card" }, { "name": "Enterprise", "price": "Custom", "note": "SSO, self-learning, on-prem or single-tenant" }, { "name": "Open source projects", "price": "Free", "note": "For OSS projects" } ], "models": [ "proprietary" ], "models_text": "Vendor-managed models; BYOK on Enterprise", "byok": null, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": true }, "summary": "Qodo (formerly CodiumAI) is a code quality platform: multi-agent PR review, a rules system that learns your standards, cross-repo impact analysis and IDE/CLI tools. Pro Team starts at $30/month with a 14-day trial; free for open-source projects.", "about": [ "Qodo 3.0 positions itself as quality control for agent-written code, including an 'Agentic Toolbox' CLI that pairs with Codex, Claude and Kiro.", "Enterprise adds BYOK, single-tenant or on-prem deployment and advanced analytics." ], "strengths": [ "Rules that learn from PR history", "Cross-repo breaking-change detection", "Free for OSS" ], "weaknesses": [ "Credit-pack pricing to understand", "Enterprise needed for SSO" ], "best_for": "Teams standardising code quality across many repos.", "alternatives": [ "coderabbit", "greptile", "graphite" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=qodo.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "amazon-q", "name": "Amazon Q Developer", "vendor": "Amazon Web Services", "url": "https://aws.amazon.com/q/developer/", "pricing_url": "https://aws.amazon.com/q/developer/pricing/", "domain": "aws.amazon.com", "status": "sunset", "status_note": "AWS blocked new Q Developer sign-ups on May 15, 2026 and will end support for the IDE plugins and paid subscriptions on April 30, 2027. The Q Developer CLI became Kiro CLI; AWS recommends migrating to Kiro.", "types": [ "extension", "cli" ], "primary": "extension", "interfaces": [ "VS Code, JetBrains, Visual Studio, Eclipse plugins", "CLI (now Kiro CLI)", "AWS Console" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 19, "tiers": [ { "name": "Free", "price": "$0 (existing users)", "note": "50 agentic requests/month" }, { "name": "Pro", "price": "$19/user/mo (existing subscriptions)", "note": "Higher limits; Java/.NET transformation" } ], "models": [ "claude" ], "models_text": "Anthropic Claude models via Amazon Bedrock", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": null, "mcp": true, "self_host": false }, "summary": "Amazon Q Developer is AWS's AI coding assistant, now being retired: new sign-ups stopped on May 15, 2026 and support ends April 30, 2027. AWS directs users to Kiro, its agentic IDE and CLI.", "about": [ "Existing Pro subscriptions ($19/user/month) keep working and can still add users until end of support. Code transformation (Java upgrades, .NET porting) also ends with the plugins.", "The newest coding models are available only in Kiro." ], "strengths": [ "Deep AWS integration (legacy)", "Java/.NET transformation features until 2027" ], "weaknesses": [ "End-of-life; no new sign-ups", "Latest models only on Kiro" ], "best_for": "Existing AWS customers finishing in-flight work before migrating to Kiro.", "alternatives": [ "kiro", "copilot", "claude-code" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=aws.amazon.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "jetbrains-ai", "name": "JetBrains AI (Junie & AI Assistant)", "vendor": "JetBrains", "url": "https://www.jetbrains.com/ai/", "pricing_url": "https://www.jetbrains.com/help/ai-assistant/licensing-and-subscriptions.html", "domain": "jetbrains.com", "status": "active", "types": [ "extension", "ide" ], "primary": "extension", "interfaces": [ "IntelliJ IDEA, PyCharm, WebStorm, GoLand, Rider and other JetBrains IDEs" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 10, "tiers": [ { "name": "AI Free", "price": "$0", "note": "Limited AI usage" }, { "name": "AI Pro", "price": "$10/mo individual ($20/user/mo org)", "note": "10 AI credits; included with All Products Pack / dotUltimate" }, { "name": "AI Ultimate", "price": "$30/mo individual ($60/user/mo org)", "note": "35 AI credits" }, { "name": "AI Enterprise", "price": "$60/user/mo", "note": "Org controls" } ], "models": [ "claude", "openai", "gemini", "local" ], "models_text": "Multiple cloud models plus local model support; 1 AI credit = $1", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": null, "pr_review": null, "mcp": true, "self_host": null }, "summary": "JetBrains AI brings AI Assistant (chat, completions) and Junie, JetBrains' coding agent, into IntelliJ-based IDEs. There is a free tier; AI Pro is $10/month and AI Ultimate $30/month for individuals.", "about": [ "Because it is built into the IDE, JetBrains AI can use the IDE's deep code understanding — inspections, refactorings and test runners — when making changes.", "AI Pro is bundled with the All Products Pack and dotUltimate subscriptions. Usage is measured in AI credits." ], "strengths": [ "Native integration with JetBrains IDE intelligence", "Junie agent included on paid tiers", "Bundled with All Products Pack" ], "weaknesses": [ "Only for JetBrains IDEs", "Credit allowances are modest on Pro" ], "best_for": "Developers who live in IntelliJ, PyCharm or other JetBrains IDEs.", "alternatives": [ "copilot", "sweep", "tabnine" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=jetbrains.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "cody", "name": "Sourcegraph Cody", "vendor": "Sourcegraph", "url": "https://sourcegraph.com/cody", "pricing_url": "https://sourcegraph.com/pricing", "domain": "sourcegraph.com", "status": "changed", "status_note": "Cody Free and Pro were discontinued on July 23, 2025. Cody is now available only to Sourcegraph Enterprise customers. Amp, Sourcegraph's agent, spun out as an independent company in December 2025.", "types": [ "extension" ], "primary": "extension", "interfaces": [ "VS Code, JetBrains", "Sourcegraph web" ], "open_source": false, "license": "Proprietary", "pricing": "enterprise", "free_tier": false, "price_from": null, "tiers": [ { "name": "Sourcegraph Enterprise", "price": "Contact sales", "note": "Cody included for Enterprise customers" } ], "models": [ "claude", "openai", "gemini" ], "models_text": "Multiple LLMs, with Sourcegraph code search as context", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": null, "self_host": true }, "summary": "Sourcegraph Cody is an AI coding assistant that uses Sourcegraph's code search for context across huge codebases. Since July 2025 it is Enterprise-only; individuals were pointed to Amp, now a separate company.", "about": [ "Cody's strength is context: it pulls symbols, APIs and usage patterns from local and remote repositories via Sourcegraph Search.", "For individual developers, Amp is the spiritual successor." ], "strengths": [ "Best-in-class cross-repo context via Sourcegraph", "Self-hosted with Sourcegraph Enterprise" ], "weaknesses": [ "No individual plans", "Requires Sourcegraph Enterprise" ], "best_for": "Large enterprises already running Sourcegraph.", "alternatives": [ "amp", "augment", "copilot" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=sourcegraph.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "phind", "name": "Phind", "vendor": "Phind (YC)", "url": "https://www.phind.com/", "pricing_url": "https://www.phind.com/", "domain": "phind.com", "status": "sunset", "status_note": "Phind shut down on January 16, 2026. Paid users received prorated refunds and data export closed on January 30, 2026.", "types": [ "other" ], "primary": "other", "interfaces": [ "Web (discontinued)" ], "open_source": false, "license": "Proprietary", "pricing": "unknown", "free_tier": null, "price_from": null, "tiers": [ { "name": "Service", "price": "Discontinued", "note": "Shut down January 16, 2026" } ], "models": [ "proprietary" ], "models_text": "Phind's own and third-party models (historically)", "byok": false, "features": { "agent": false, "autocomplete": false, "cloud_agents": false, "pr_review": false, "mcp": false, "self_host": false }, "summary": "Phind was an AI search engine and answer assistant for developers. It shut down on January 16, 2026; use ChatGPT, Claude or Perplexity-style tools, or an IDE assistant, instead.", "about": [ "Phind combined web search with coding-tuned models to answer technical questions with sources.", "The service is no longer available." ], "strengths": [ "(Historical) fast, cited technical answers" ], "weaknesses": [ "Discontinued" ], "best_for": "—", "alternatives": [ "copilot", "claude-code", "cursor" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=phind.com&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "blackbox", "name": "Blackbox AI", "vendor": "Blackbox AI", "url": "https://www.blackbox.ai/", "pricing_url": "https://www.blackbox.ai/pricing", "domain": "blackbox.ai", "status": "changed", "status_note": "Blackbox's site now centres on a model platform: a 300+ model catalog, router, dedicated deployments and per-token pricing with enterprise commitments, alongside its coding agents.", "types": [ "other", "cloud" ], "primary": "other", "interfaces": [ "Web", "API", "IDE extension (check site)" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": null, "price_from": null, "tiers": [ { "name": "Pay per token", "price": "Model rack rates", "note": "300+ models" }, { "name": "Enterprise", "price": "Custom (annual commit)", "note": "Dedicated deployment, SSO, cloud coding agents API" } ], "models": [ "claude", "openai", "gemini", "open" ], "models_text": "300+ models via its catalog and router", "byok": true, "features": { "agent": true, "autocomplete": null, "cloud_agents": true, "pr_review": null, "mcp": null, "self_host": true }, "summary": "Blackbox AI started as a coding assistant and now primarily sells a model platform — a 300+ model catalog with routing, dedicated deployments and per-token pricing — plus cloud coding agents via API. Check the official site for current consumer plans.", "about": [ "Blackbox publishes its own Terminal-Bench runs and offers remote agent runners and an Agents API for cloud coding agents on Enterprise.", "Pricing is per token at catalog rates, with discounts for committed annual spend." ], "strengths": [ "Huge model catalog behind one API", "Enterprise single-tenant deployment" ], "weaknesses": [ "Positioning has shifted; consumer plan details unclear", "Hard to compare with seat-based tools" ], "best_for": "Teams that want one API across many models plus agent infrastructure.", "alternatives": [ "opencode", "kilo-code", "factory" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=blackbox.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": false }, { "slug": "bito", "name": "Bito", "vendor": "Bito", "url": "https://bito.ai/", "pricing_url": "https://bito.ai/pricing/", "domain": "bito.ai", "status": "active", "types": [ "review", "other" ], "primary": "review", "interfaces": [ "GitHub, GitLab, Bitbucket (PR reviews)", "IDE", "Drop-in endpoint for Claude Code, Cursor, Codex, Copilot" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 15, "tiers": [ { "name": "AI Code Reviews — Team", "price": "$15/seat/mo ($12 annual)", "note": "5K lines/seat/month, then $5 per 1K" }, { "name": "AI Code Reviews — Professional", "price": "$25/seat/mo ($20 annual)", "note": "Custom guidelines, CI/CD reviews; free trial" }, { "name": "Governor / AI Architect", "price": "Usage-based", "note": "Model routing and context engine; contact sales" }, { "name": "Enterprise", "price": "Contact us", "note": "Self-hosted option" } ], "models": [ "claude", "openai", "gemini", "grok", "open" ], "models_text": "Frontier models from Anthropic, OpenAI, Google, xAI plus open-weight models; BYOK supported", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": false, "pr_review": true, "mcp": null, "self_host": true }, "summary": "Bito sells AI code reviews ($15/seat/month Team, $25 Professional) plus 'Governor', a model router and context engine that claims to cut coding-agent spend for tools like Claude Code, Cursor and Codex.", "about": [ "Bito's code reviewer comments on pull requests with codebase context and can learn from feedback; Professional adds CI/CD pipeline reviews.", "Its newer products sit in front of your existing agents, routing requests across frontier and open-weight models." ], "strengths": [ "Low-cost per-seat code review", "Works with agents you already use", "Self-hosted option" ], "weaknesses": [ "Line-based overage pricing", "Savings claims are vendor-reported" ], "best_for": "Teams wanting affordable PR review or tighter control of agent model costs.", "alternatives": [ "coderabbit", "greptile", "qodo" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=bito.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "askcodi", "name": "AskCodi", "vendor": "AskCodi", "url": "https://www.askcodi.com/", "pricing_url": "https://www.askcodi.com/pricing", "domain": "askcodi.com", "status": "changed", "status_note": "AskCodi has repositioned from an IDE coding assistant to an 'agency' that scopes, builds and maintains apps, plus an AI gateway, priced in tokens.", "types": [ "builder", "other" ], "primary": "builder", "interfaces": [ "Web", "Desktop app", "AI Gateway API" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 29, "tiers": [ { "name": "Free", "price": "$0", "note": "1M tokens/month (300K/day), one app" }, { "name": "Individual", "price": "$29/mo", "note": "Full price returned as tokens; one app" }, { "name": "Pro", "price": "$99/mo", "note": "Several apps" }, { "name": "Business", "price": "$199/mo", "note": "A portfolio of apps" }, { "name": "Enterprise", "price": "Talk to sales", "note": "Up to 500 seats" } ], "models": [ "claude", "openai", "gemini", "open" ], "models_text": "Multiple models through its AI Gateway", "byok": null, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": false, "mcp": null, "self_host": false }, "summary": "AskCodi is now an AI 'agency' that builds and looks after websites and apps, plus an AI model gateway, with token-based plans from free (1M tokens/month) to $29, $99 and $199 per month.", "about": [ "Every paid plan returns its full price as tokens; plans differ by how many apps you run.", "If you want a classic IDE assistant, see Copilot, Cline or Kilo Code instead." ], "strengths": [ "Simple token-back pricing", "Free tier with no card" ], "weaknesses": [ "No longer a typical IDE coding assistant" ], "best_for": "Small businesses wanting an AI-built and maintained app.", "alternatives": [ "lovable", "replit", "bolt" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=askcodi.com&sz=128", "yc": false, "benchmarks": [], "recommendable": false }, { "slug": "refact", "name": "Refact.ai", "vendor": "Small Magellanic Cloud AI Ltd", "url": "https://refact.ai/", "pricing_url": "https://refact.ai/", "domain": "refact.ai", "status": "sunset", "status_note": "Refact.ai's website banner states that Refact Cloud is shutting down; self-hosted and enterprise offerings are still listed. Check the official site before adopting.", "types": [ "extension" ], "primary": "extension", "interfaces": [ "VS Code, JetBrains plugins", "Self-hosted server" ], "open_source": true, "license": "Open source (self-hosted server)", "pricing": "enterprise", "free_tier": null, "price_from": null, "tiers": [ { "name": "Refact Cloud", "price": "Shutting down", "note": "See official announcement" }, { "name": "Self-hosted / Enterprise", "price": "Contact vendor", "note": "Check official site" } ], "models": [ "open", "local" ], "models_text": "Self-hosted open models (and cloud models historically)", "byok": true, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": null, "self_host": true }, "summary": "Refact.ai is an open-source AI coding agent and autocomplete you can self-host. Its hosted Refact Cloud is shutting down, so it is now mainly relevant for self-hosted enterprise setups.", "about": [ "Refact offered an autonomous agent plus completions in VS Code and JetBrains, with a self-hostable server for running models on your own GPUs.", "Because the cloud service is winding down, new users should verify the current status first." ], "strengths": [ "Self-hostable, open models" ], "weaknesses": [ "Cloud service shutting down", "Uncertain roadmap" ], "best_for": "Teams that must run everything on their own hardware.", "alternatives": [ "tabnine", "continue", "openhands" ], "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=refact.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "pieces", "name": "Pieces", "vendor": "Pieces for Developers", "url": "https://pieces.app/", "pricing_url": "https://pieces.app/pricing", "domain": "pieces.app", "status": "active", "types": [ "other" ], "primary": "other", "interfaces": [ "Desktop app", "MCP server for other AI tools", "IDE & browser extensions" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": null, "price_from": 18.99, "tiers": [ { "name": "Pro", "price": "$18.99/user/mo (save 25% yearly)", "note": "Personal memory across tools; trial requires card" }, { "name": "Enterprise", "price": "$22.99/user/mo", "note": "Org-wide controls, approved AI providers, own keys" } ], "models": [ "claude", "openai", "gemini", "local" ], "models_text": "On-device and cloud models; Enterprise can use approved providers and your own API keys", "byok": true, "features": { "agent": false, "autocomplete": false, "cloud_agents": false, "pr_review": false, "mcp": true, "self_host": false }, "summary": "Pieces is a long-term memory layer for developers: it captures context from your workflow on-device and exposes it to any MCP-ready AI tool. Pro is $18.99/user/month; it complements, rather than replaces, a coding agent.", "about": [ "Pieces remembers snippets, links and work context across your IDE, browser and chats, then lets assistants like Claude Code or Cursor query that history via MCP.", "Processing is designed to run on-device where possible." ], "strengths": [ "Gives any agent memory of your past work", "MCP integration", "On-device focus" ], "weaknesses": [ "Not a coding agent by itself", "Paid-only for full features" ], "best_for": "Developers who want their AI tools to remember context across sessions.", "alternatives": [ "claude-code", "cursor" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=pieces.app&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "zencoder", "name": "Zencoder", "vendor": "Zencoder", "url": "https://zencoder.ai/", "pricing_url": "https://zencoder.ai/pricing", "domain": "zencoder.ai", "status": "active", "types": [ "extension", "cloud" ], "primary": "extension", "interfaces": [ "VS Code & JetBrains agents", "Zenflow (web)" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 45, "tiers": [ { "name": "Pro", "price": "$45/user/mo ($40 annual)", "note": "30,000 monthly credits; 7-day trial" }, { "name": "Pro Plus", "price": "$95/user/mo ($85 annual)", "note": "" }, { "name": "Pro Max", "price": "$195/user/mo ($175 annual)", "note": "" }, { "name": "Enterprise", "price": "Custom", "note": "" } ], "models": [ "claude", "openai", "gemini" ], "models_text": "Multiple frontier models (check site)", "byok": null, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": null, "mcp": true, "self_host": null }, "summary": "Zencoder provides AI coding agents for VS Code and JetBrains plus Zenflow, an orchestration layer for multi-agent workflows aimed at engineering managers. Pro starts at $45/user/month with a 7-day trial.", "about": [ "Zencoder's IDE agents use 'repo grokking' to build codebase context; Zenflow coordinates agents across tasks and reviews.", "Plans are credit-based per user." ], "strengths": [ "VS Code and JetBrains coverage", "Multi-agent orchestration" ], "weaknesses": [ "Expensive vs. Copilot/Cursor", "No free tier" ], "best_for": "Teams standardising on agent workflows across IDEs.", "alternatives": [ "augment", "copilot", "kiro" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=zencoder.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "augment", "name": "Augment Code", "vendor": "Augment Code", "url": "https://www.augmentcode.com/", "pricing_url": "https://www.augmentcode.com/pricing", "domain": "augmentcode.com", "status": "active", "types": [ "extension", "cli", "cloud" ], "primary": "extension", "interfaces": [ "VS Code & JetBrains", "CLI (Auggie)", "Cosmos (cloud)" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": false, "price_from": 20, "tiers": [ { "name": "Standard", "price": "$20/mo flat", "note": "$20 of usage; up to 50 seats included; no AI training" }, { "name": "Business", "price": "$100/mo flat", "note": "$100 of usage; SSO, audit trails" }, { "name": "Enterprise", "price": "Custom", "note": "Custom usage and security" } ], "models": [ "claude", "openai" ], "models_text": "Frontier models with Augment's Context Engine", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": true, "pr_review": true, "mcp": true, "self_host": false }, "summary": "Augment Code is an AI coding agent for large codebases, built around its Context Engine, available in VS Code, JetBrains and a CLI. Plans are flat-rate rather than per seat: Standard is $20/month and Business $100/month, each including up to 50 seats.", "about": [ "Augment indexes your entire codebase in real time so its agent and completions can reason across services and repos.", "Usage across LLM, Context Engine and compute draws from the included monthly amount, with pay-as-you-go top-ups." ], "strengths": [ "Strong large-codebase context", "Unusual flat team pricing", "No training on your code" ], "weaknesses": [ "Usage-metered beyond the included amount", "Closed source" ], "best_for": "Teams working in big, multi-repo codebases.", "alternatives": [ "cursor", "cody", "copilot" ], "status_note": null, "pricing_note": null, "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=augmentcode.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "sweep", "name": "Sweep", "vendor": "Sweep AI (YC)", "url": "https://sweep.dev/", "pricing_url": "https://sweep.dev/pricing", "domain": "sweep.dev", "status": "active", "status_note": "Sweep began as a GitHub bot that turned issues into PRs; it is now an AI assistant plugin for JetBrains IDEs.", "types": [ "extension" ], "primary": "extension", "interfaces": [ "JetBrains IDEs plugin" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 10, "tiers": [ { "name": "Free trial", "price": "$0", "note": "1,000 autocompletes + $5 API credits" }, { "name": "Basic", "price": "$10/mo", "note": "Unlimited autocomplete" }, { "name": "Pro", "price": "$20/mo", "note": "" }, { "name": "Ultra", "price": "$60/mo", "note": "Agent usage at model cost + 5%" } ], "pricing_note": "Prices from sweep.dev/pricing as surfaced by search (page did not render for our fetcher) — confirm on the official site.", "models": [ "claude", "openai" ], "models_text": "Hosted models; agent usage billed at underlying model cost + 5%", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": false, "pr_review": false, "mcp": null, "self_host": false }, "summary": "Sweep is an AI coding assistant built specifically for JetBrains IDEs, with fast autocomplete, chat and an agent. Plans run from $10 to $60/month after a free trial.", "about": [ "Sweep focuses on making JetBrains users feel first-class: next-edit autocomplete tuned for IntelliJ-family IDEs plus chat and agent modes.", "Chat and code generation consume API credits that roll over for 30 days." ], "strengths": [ "Purpose-built for JetBrains", "Cheap entry plan" ], "weaknesses": [ "JetBrains only", "Smaller vendor" ], "best_for": "IntelliJ/PyCharm users who want Cursor-like autocomplete.", "alternatives": [ "jetbrains-ai", "copilot", "tabnine" ], "new": false, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=sweep.dev&sz=128", "yc": true, "benchmarks": [], "recommendable": true }, { "slug": "kiro", "name": "Kiro", "vendor": "Amazon Web Services", "url": "https://kiro.dev/", "pricing_url": "https://kiro.dev/pricing/", "domain": "kiro.dev", "status": "active", "new": true, "types": [ "ide", "cli" ], "primary": "ide", "interfaces": [ "Kiro IDE (VS Code-based)", "Kiro CLI" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Free", "price": "$0", "note": "Starter credits" }, { "name": "Pro", "price": "$20/user/mo", "note": "1,000 credits" }, { "name": "Pro+", "price": "$40/user/mo", "note": "2,000 credits" }, { "name": "Pro Max", "price": "$100/user/mo", "note": "" }, { "name": "Power", "price": "$200/user/mo", "note": "" }, { "name": "Add-on credits", "price": "$0.04/credit", "note": "" } ], "models": [ "claude" ], "models_text": "Anthropic Claude models plus an 'Auto' router", "byok": false, "features": { "agent": true, "autocomplete": true, "cloud_agents": null, "pr_review": null, "mcp": true, "self_host": false }, "summary": "Kiro is AWS's agentic IDE and CLI, known for spec-driven development: it turns a prompt into requirements, a design and a task list before writing code. Free tier available; Pro is $20/user/month. It replaces Amazon Q Developer.", "about": [ "Kiro adds steering files, event-driven hooks, custom subagents and 'Powers' on top of a VS Code-based editor, so VS Code settings and extensions import directly.", "The Kiro CLI is the successor to the Amazon Q Developer CLI and stays backwards compatible with q commands." ], "strengths": [ "Spec-driven workflow keeps agents on track", "Hooks and steering for team conventions", "Official migration path from Amazon Q" ], "weaknesses": [ "Credit-based limits", "Mostly Claude models" ], "best_for": "Teams that want structured, spec-first agent development — especially AWS shops.", "alternatives": [ "cursor", "windsurf", "claude-code" ], "status_note": null, "pricing_note": null, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=kiro.dev&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "amp", "name": "Amp", "vendor": "Amp Inc. (spun out of Sourcegraph)", "url": "https://ampcode.com/", "pricing_url": "https://ampcode.com/pricing", "domain": "ampcode.com", "status": "active", "new": true, "types": [ "cli", "cloud" ], "primary": "cli", "interfaces": [ "Terminal (Amp CLI)", "Remote 'orbs' (cloud runners)", "Editor integrations" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": 20, "tiers": [ { "name": "Hobby", "price": "Free", "note": "All features; BYOK or your ChatGPT/Claude subscriptions; pay-as-you-go orbs" }, { "name": "Megawatt", "price": "$20/mo", "note": "45,000 orb minutes" }, { "name": "Gigawatt", "price": "Higher tier", "note": "Check official site" }, { "name": "Teams", "price": "No extra charge", "note": "Shared threads, SSO, pooled credits" }, { "name": "Enterprise", "price": "Custom", "note": "SCIM, audit logs" } ], "models": [ "claude", "openai", "gemini" ], "models_text": "Frontier models; use your own keys or existing ChatGPT/Claude subscriptions with no Amp token fees", "byok": true, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": null, "mcp": true, "self_host": false }, "summary": "Amp is a frontier coding agent from the team that built Sourcegraph, now an independent company. It runs in the terminal and on remote 'orbs', and its free Hobby plan lets you use your own keys or ChatGPT/Claude subscriptions with no Amp token fees.", "about": [ "Amp emphasises multi-threaded agent work, shared team threads and plugins. Remote orbs give agents their own machines, billed per minute when you need them.", "It is the recommended successor for former Cody Free/Pro users." ], "strengths": [ "Use existing subscriptions or keys, no token markup", "Cloud runners on demand", "Strong team collaboration features" ], "weaknesses": [ "Orb-minute pricing is novel", "Fast-moving product" ], "best_for": "Agent power users who want flexibility over models and where agents run.", "alternatives": [ "claude-code", "codex", "factory" ], "status_note": null, "pricing_note": null, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=ampcode.com&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "factory", "name": "Factory (Droids)", "vendor": "Factory", "url": "https://factory.ai/", "pricing_url": "https://factory.ai/pricing", "domain": "factory.ai", "status": "active", "new": true, "types": [ "cli", "cloud" ], "primary": "cli", "interfaces": [ "Terminal (Droid CLI)", "Web", "IDE" ], "open_source": false, "license": "Proprietary", "pricing": "paid", "free_tier": null, "price_from": 20, "tiers": [ { "name": "Pro", "price": "$20/mo", "note": "" }, { "name": "Plus", "price": "$100/mo", "note": "" }, { "name": "Max", "price": "$200/mo", "note": "" }, { "name": "Teams", "price": "$60/mo per team + $40/mo per seat", "note": "Shared billing" }, { "name": "Enterprise", "price": "Contact sales", "note": "" } ], "models": [ "claude", "openai", "gemini", "open" ], "models_text": "All leading frontier and open-weight models (GPT, Claude, Gemini and more)", "byok": null, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": null, "mcp": true, "self_host": null }, "summary": "Factory builds 'Droids' — model-agnostic coding agents that run in your terminal, IDE or the cloud and handle tasks from feature work to migrations. Plans start at $20/month; Teams are $60/month plus $40 per seat.", "about": [ "Droids support all major frontier and open-weight models, so you can pick per task. Factory targets enterprise-scale engineering work as well as individual developers.", "Check the official site for included usage on each tier." ], "strengths": [ "Model-agnostic", "Terminal, IDE and cloud from one product" ], "weaknesses": [ "Usage limits per tier not obvious", "Newer vendor" ], "best_for": "Developers who want a model-agnostic Claude Code alternative with cloud options.", "alternatives": [ "claude-code", "amp", "codex" ], "status_note": null, "pricing_note": null, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=factory.ai&sz=128", "yc": false, "benchmarks": [], "recommendable": true }, { "slug": "jules", "name": "Jules", "vendor": "Google", "url": "https://jules.google/", "pricing_url": "https://jules.google/", "domain": "jules.google", "status": "active", "new": true, "types": [ "cloud" ], "primary": "cloud", "interfaces": [ "Web", "GitHub (issue label)", "CLI" ], "open_source": false, "license": "Proprietary", "pricing": "freemium", "free_tier": true, "price_from": null, "tiers": [ { "name": "Jules (free)", "price": "$0", "note": "15 tasks/day, 3 concurrent" }, { "name": "Pro", "price": "Via Google AI Pro", "note": "100 tasks/day, latest models" }, { "name": "Ultra", "price": "Via Google AI Ultra", "note": "300 tasks/day, priority access" } ], "models": [ "gemini" ], "models_text": "Google Gemini models (Gemini 3 Pro on paid tiers)", "byok": false, "features": { "agent": true, "autocomplete": false, "cloud_agents": true, "pr_review": null, "mcp": null, "self_host": false }, "summary": "Jules is Google's asynchronous coding agent: pick a GitHub repo and branch, describe the task, and it clones the code to a cloud VM, plans, makes the change and opens a PR. Free plan includes 15 tasks per day.", "about": [ "Jules shows its plan before executing and returns a diff for approval. You can also assign tasks by adding a 'jules' label to a GitHub issue.", "Higher daily task limits come with Google AI Pro and Ultra subscriptions." ], "strengths": [ "Generous free daily task quota", "Simple GitHub-first workflow" ], "weaknesses": [ "Gemini models only", "GitHub only" ], "best_for": "Developers who want to offload chores (dependency bumps, tests, bug fixes) to a background agent for free.", "alternatives": [ "codex", "devin", "copilot" ], "status_note": null, "pricing_note": null, "last_verified": "2026-10-10", "favicon": "https://www.google.com/s2/favicons?domain=jules.google&sz=128", "yc": false, "benchmarks": [], "recommendable": true } ]