Ranked · Coding agents
The Best AI Coding Agents (2026)
25 AI coding agents compared on the split that actually matters — in-editor GUI vs. terminal-native vs. fully autonomous cloud vs. browser-based vs. open-source/self-hosted vs. automated code review — not a single forced ranking.
The ranking
26 agents, compared. Start with fit and price, then open the full assessment. Payment does not alter this ranking. Selection and method ↓
All picks at a glance
Compare fit and price, then choose a name to read the full assessment.
| # | Agent | Best for | From |
|---|---|---|---|
| 1 | CursorIn-editor IDE agentsvs GitHub Copilot | Professional developers who want a powerful AI agent woven into a familiar VS Code-based editor, with first-class model choice and the option to be hands-off via Cursor Router. | Free |
| 2 | GitHub CopilotIn-editor IDE agentsvs Cursor | Enterprise and team developers already in GitHub and mainstream IDEs who want governed, reviewable AI coding at scale, with optional access to Claude Code and OpenAI Codex through the same interface. | Free |
| 3 | Devin DesktopIn-editor IDE agentsvs Devin | Developers who want a full, agent-native desktop IDE and the freedom to run several coding agents, including third-party ones, side by side. | Free |
| 4 | ZedIn-editor IDE agents | Developers who want a fast, open-source editor and the freedom to run Claude Code, Codex, Gemini CLI or other vendors' own agents natively via the Agent Client Protocol Zed created, instead of being locked into one editor's built-in agent. | Free |
| 5 | Claude CodeTerminal-native agentsvs Cursor | Developers and teams who want a terminal-native, scriptable agent that composes with CI and real engineering workflows. | Free |
| 6 | Codex CLITerminal-native agents | Developers who already pay for a ChatGPT plan and want a terminal-native agent with granular, inspectable sandbox/permission controls across CLI, desktop and IDE surfaces. | Free |
| 7 | Google Antigravity CLITerminal-native agents | Developers already inside the Google ecosystem (Google AI subscription, GCP billing) who want a free terminal agent with Gemini, Claude and gpt-oss model access, especially anyone migrating off a now-sunset Gemini CLI workflow. | Freemium |
| 8 | WarpTerminal-native agents | Developers and teams who want one open platform that runs and coordinates multiple coding agents (Claude Code, Codex, Gemini CLI, Warp Agent) side by side across terminal, CLI and cloud, with shared team context and programmatic API access to cloud agent runs. | Free |
| 9 | DevinFully autonomous cloud agentsvs Devin Desktop | Engineering teams offloading well-scoped migration tasks without adding headcount | Free |
| 10 | Replit AgentFully autonomous cloud agents | Non-technical founders who want a chat prompt to become a deployed, database-backed app | Free |
| 11 | FactoryFully autonomous cloud agents | Engineering organizations that want AI agents handling more of the SDLC than code generation alone (triage, testing, release and documentation) and are willing to pay enterprise pricing for deployment flexibility and compliance controls. | $20/mo |
| 12 | BoltBrowser-based app builders | Founders, designers and developers who want a working, deployed full-stack app straight from a prompt, Figma frame or Google Stitch design, with zero local setup. | Free |
| 13 | ClineOpen-source & BYOK agentsvs Cursor | Developers and teams who want a transparent, open-source, model-agnostic agent they fully control and pay for by usage, with optional flat-rate access to open-weight models. | Free |
| 14 | AiderOpen-source & BYOK agents | Developers who want free, model-agnostic AI pair programming in their terminal | Free |
| 15 | OpenCodeOpen-source & BYOK agents | Developers who want a genuinely free, fully open-source coding agent that works with any model provider, including their own local Ollama models, without being locked into one vendor's terminal tool. | Free |
| 16 | OpenHandsOpen-source & BYOK agents | Open source-minded engineering teams and enterprises that want a self-hostable coding agent platform | Free |
| 17 | JugglerNew open-source session-tree agentsvs zot | Developers who want hands-on visibility into what an AI coding agent is doing to their codebase and accept early-stage, self-hosted, open-source software. | Free |
| 18 | zotNew open-source session-tree agentsvs Juggler | Developers who want a fast, free, Go-native coding-agent harness with real session-branching control and are comfortable running fast-moving, self-described "beta forever" software from an independent maintainer. | Free |
| 19 | CodeRabbitCode review agents | Engineering teams, using a coding agent or writing code by hand, that want automatic, consistent PR review plus triage, explanation and continuous security across GitHub, GitLab, Bitbucket or Azure DevOps before a human reviewer looks at it. | Free |
| 20 | Muse CodeTerminal-native agents | Cost-sensitive teams willing to trade code privacy for Contributor tier's roughly 12x cheaper input and 21x cheaper output, or anyone with a large multi-part refactor that benefits from several sub-agents working in parallel isolated worktrees. | Usage-based |
| 21 | JunieIn-editor IDE agents | Developers and teams already standardized on a JetBrains IDE who want an agent that shares the same debugger, project index, and version control already built into their editor. | Free |
| 22 | Augment CodeFully autonomous cloud agents | Engineering leaders at organizations who want AI agents running full SDLC loops end to end, with humans at the quality gates, rather than inline completion assistants. | $100/mo |
| 23 | Gemini Code AssistIn-editor IDE agents | Teams standardized on VS Code or JetBrains who are already on Google Cloud | $19/seat/yr |
| 24 | gooseOpen-source & BYOK agents | Devs who want an open source coding agent that reuses an existing Claude or ChatGPT plan | Free |
| 25 | Kilo CodeOpen-source & BYOK agents | Free | |
| 26 | LettaIn this category, not yet ranked | Developers and researchers who specifically need an agent whose memory, skills and identity genuinely persist and evolve across long-running sessions, rather than reset on every call. | Free |
Selection, context, and ranking method
Coding agents split first on where you use them, not on which one is "best." Cursor, GitHub Copilot, Windsurf (now Cognition's Devin Desktop), Zed, Junie and Gemini Code Assist put the agent inside a familiar graphical editor. Zed is the only one of the six built from scratch (GPUI/Rust) rather than forked from VS Code, Junie is JetBrains' own and the deepest inside the IDEs it owns, and Gemini Code Assist is Google's — running its Gemini 3 model with a 1M-token context window across VS Code, the full JetBrains family and Google's own Cloud IDEs.
Claude Code, OpenAI Codex CLI, Google's Antigravity CLI, Meta's Muse Code and Warp put it in the terminal instead, for developers who want it to compose with scripts and CI. Antigravity CLI replaced the open-source Gemini CLI in June 2026, Muse Code launched in early August 2026 with a steep Contributor-tier discount in exchange for Meta training on your prompts and code, and Warp takes a different approach entirely: not a CLI tool but a full terminal application that now also ships its own standalone coding agent and orchestrates Claude Code, Codex, Gemini CLI and Warp's own agent side by side inside it, rather than being one more agent to choose between.
Devin, Replit Agent, Factory and Augment Code go further still, running asynchronously so you delegate a whole ticket, app, an entire engineering pipeline of specialized "Droid" agents, or Augment Code's full-loop Cosmos orchestration layer for code review, ticket-to-PR, security remediation and incident response, rather than pair with it live. Bolt.new skips local setup entirely, building and deploying an app from a prompt inside the browser tab itself, using its own WebContainers runtime rather than a remote VM.
A fourth group of self-hosted, BYOK agents — Cline, Aider, OpenCode, OpenHands, Goose and Kilo Code — trades a subscription for full control: from mature and widely adopted (Cline, Aider, OpenCode) to built for enterprise self-hosting (OpenHands, which layers SSO/RBAC/audit logging onto its MIT-licensed core) to Linux Foundation-governed with a subscription-reuse option built in (Goose, whose ACP support lets it run against an existing Claude, ChatGPT or Gemini plan instead of a separate API key) to MIT-licensed free-to-self-host with a 500+ model hosted Gateway as an alternative to pure BYOK (Kilo Code).
A fifth group (Juggler and Zot) takes a genuinely different approach to the session model itself, branching a conversation into a tree instead of scrolling a linear transcript. CodeRabbit sits outside all of this: instead of writing code, it automatically reviews the pull requests every other tool on this page produces, pairing with a coding agent instead of competing with one.
For most professional developers who want the smoothest experience with the least setup, Cursor remains the category's default pick. But the honest answer to "which is best" depends entirely on whether you want a subscription-priced GUI, a scriptable terminal agent, a fire-and-forget cloud engineer or full-SDLC platform, a zero-setup browser app builder, full ownership of your own keys and models, or an automatic reviewer for the code any of the above produce.
Every entry below is a Published listing that already cleared this site's per-listing quality gate: sourced pricing verified live, admitted weaknesses, a stated "best for" and "not for," and a rolling re-verification cadence (see /methodology). coding-agents is the site's largest category by both listing count (25) and traffic. Its own category page and several of the site's tool-specific "alternatives" pages (including Cursor, Claude Code, GitHub Copilot and Windsurf) already live here without a single page tying them together, so this ranking closes a real gap rather than opening one.
It also runs alongside the flagship /best-ai-agents ranking's own "Coding agents" group, which lists all 25 of the same listings in one flat list. This page's groupings (interface and business model, not a single forced order) are what that flat list can't give you. The grouping below follows the real, buyer-relevant split in how each tool is used: IDE-embedded, terminal-native, fully autonomous cloud, browser-based, open-source/self-hosted, or automated review rather than authoring. It does not force a single ranking on tools that serve genuinely different jobs. Live score chips (evidence, honesty, depth, freshness) are computed the same way as each tool's own listing page.
Muse Code, Meta's own terminal-native agent, joined the category on August 9, 2026, as the 20th entry, grouped with Terminal-native agents. Junie, JetBrains' own IDE-embedded coding agent, joined on August 10, 2026 as the 21st entry, grouped with In-editor IDE agents alongside Cursor and GitHub Copilot. Augment Code, an agent platform built around its Cosmos orchestration engine for full-loop code review, ticket-to-PR delivery, security remediation and incident response, joined on August 16, 2026 as the 22nd entry, grouped with Fully autonomous cloud agents. Gemini Code Assist, Google's own IDE-embedded coding assistant, joined on August 16, 2026 as the 23rd entry, grouped with In-editor IDE agents alongside Cursor, GitHub Copilot, Windsurf, Zed and Junie.
Goose, an open-source coding agent now governed by the Linux Foundation's Agentic AI Foundation, joined on August 17, 2026 as the 24th entry, grouped with Open-source & BYOK agents alongside Cline, Aider, OpenCode and OpenHands. Kilo Code, an open-source (MIT), self-hostable coding agent whose hosted Gateway reaches 500+ models at provider rates, joined on August 18, 2026 as the 25th entry, grouped with Open-source & BYOK agents alongside Cline, Aider, OpenCode, OpenHands and Goose. As of 26 August 2026 no further coding-agent listing has joined that clears this page's bar; the most recent addition, Kilo Code, has held the 25th slot since 18 August and the page still ranks all 25 Published entries alive on the site today.
The ranking
In-editor IDE agents
Editor’s pick
#1
Cursor
Coding agent in a desktop editor, terminal CLI, cloud VMs and CI, priced per seat plus token usage.
Best forProfessional developers who want a powerful AI agent woven into a familiar VS Code-based editor, with first-class model choice and the option to be hands-off via Cursor Router.
The category’s adoption leader for in-editor agent work: fast multi-file edits and per-request model choice (Claude, GPT, Gemini), at the cost of closed-source lock-in and Pro-tier fast-request limits under heavy use.
- #2
GitHub Copilot
GitHub's AI coding assistant and cloud agent, from editor completions to pull-request review.
Best forEnterprise and team developers already in GitHub and mainstream IDEs who want governed, reviewable AI coding at scale, with optional access to Claude Code and OpenAI Codex through the same interface.
The safest enterprise default: the widest IDE coverage, a native GitHub issue-to-PR agent mode and the most mature governance controls, deliberately more conservative than Cursor or Devin.
- #3
Devin Desktop
AI IDE with a Kanban command center for running local and cloud coding agents; the renamed Windsurf.
Best forDevelopers who want a full, agent-native desktop IDE and the freedom to run several coding agents, including third-party ones, side by side.
The former Windsurf, now Cognition’s Devin Desktop: a vendor-neutral "agent command centre" that runs Claude, Codex and Gemini side by side via the open Agent Client Protocol, if you can stomach three rebrands in two years.
- #4
Zed
Rust code editor with an Agent Panel running Zed's own agent, ACP external agents, or terminal agent CLIs.
Best forDevelopers who want a fast, open-source editor and the freedom to run Claude Code, Codex, Gemini CLI or other vendors' own agents natively via the Agent Client Protocol Zed created, instead of being locked into one editor's built-in agent.
The only editor here built from scratch (GPUI/Rust) instead of forked from VS Code: open source and the creator of the Agent Client Protocol that runs Claude Code, Codex and Gemini CLI natively, now adopted by JetBrains too, but its extensions and remote-dev story are still catching up to VS Code's.
Terminal-native agents
- #5
Claude Code
Agentic coding tool from Anthropic that edits your codebase, runs commands, and opens pull requests.
Best forDevelopers and teams who want a terminal-native, scriptable agent that composes with CI and real engineering workflows.
The strongest command-line-first agent for whole-repository reasoning: scriptable into CI, included from the $17/month Claude Pro plan, with usage-based cost the thing to watch on long runs.
- #6
Codex CLI
OpenAI's terminal coding agent for inspecting, editing and running repository code from the shell.
Best forDevelopers who already pay for a ChatGPT plan and want a terminal-native agent with granular, inspectable sandbox/permission controls across CLI, desktop and IDE surfaces.
OpenAI’s open-source (Apache-2.0) terminal agent, bundled into every ChatGPT plan from Free to Enterprise instead of sold separately: the catch is real-world cost (OpenAI’s own numbers put regular use at $100–200/month) and a documented June 2026 rate-limit regression.
- #7
Google Antigravity CLI
Terminal agent from Google Antigravity: an agy TUI plus headless mode for macOS, Linux and Windows.
Best forDevelopers already inside the Google ecosystem (Google AI subscription, GCP billing) who want a free terminal agent with Gemini, Claude and gpt-oss model access, especially anyone migrating off a now-sunset Gemini CLI workflow.
Google's forced replacement for the sunset Gemini CLI: a genuinely generous free tier bundling both Gemini and Claude models, with real concurrent-subagent execution, but closed-source and not yet at feature parity with the tool it replaced.
- #8
Warp
Agentic development environment: a terminal, a standalone agent CLI, and unattended cloud coding agents.
Best forDevelopers and teams who want one open platform that runs and coordinates multiple coding agents (Claude Code, Codex, Gemini CLI, Warp Agent) side by side across terminal, CLI and cloud, with shared team context and programmatic API access to cloud agent runs.
Not another single-vendor CLI but the terminal itself, rebuilt to orchestrate Claude Code and Codex side by side: open source (dual AGPL-3.0/MIT) since April 2026, with a separate Warp Factories cloud-agent orchestration platform now in Early Access; adopting the terminal piece still means switching your whole terminal, not just adding a tool.
Fully autonomous cloud agents
- #9
Devin
Cloud AI software engineer that takes a ticket, works in its own VM, and opens a pull request.
Best forEngineering teams offloading well-scoped migration tasks without adding headcount
The boldest bet on full autonomy: delegate a ticket and get back a reviewed pull request, with a fleet of parallel Devins for large migrations; independent reviewers still find its autonomy uneven on ambiguous tasks, so treat its PRs as drafts.
- #10
Replit Agent
AI agent that writes, tests and publishes full apps from a chat prompt in Replit's browser workspace.
Best forNon-technical founders who want a chat prompt to become a deployed, database-backed app
The lowest-friction path from a prompt to a live, deployed full-stack app (no local setup at all), but effort-based credit billing can escalate fast on real feature work.
- #11
Factory
Droid coding agents that run in the desktop app, CLI, IDEs, CI and Factory-managed cloud computers.
Best forEngineering organizations that want AI agents handling more of the SDLC than code generation alone (triage, testing, release and documentation) and are willing to pay enterprise pricing for deployment flexibility and compliance controls.
A full-SDLC bet: coordinated specialist "Droid" agents (code, review, test, docs, monitor) with real enterprise traction behind it ($150M Series C at a $1.5B valuation), but no free tier and its deepest controls are sales-gated.
Browser-based app builders
Best forFounders, designers and developers who want a working, deployed full-stack app straight from a prompt, Figma frame or Google Stitch design, with zero local setup.
Runs a genuine Node.js environment natively inside the browser tab (WebContainers, not a remote VM) for instant previews: a real, well-funded technical edge, offset by a poor 1.5/5 Trustpilot support record and no public API.
Open-source & BYOK agents
- #13
Cline
Apache 2.0 coding agent for IDE, terminal and SDK, running on any model provider you choose.
Best forDevelopers and teams who want a transparent, open-source, model-agnostic agent they fully control and pay for by usage, with optional flat-rate access to open-weight models.
The open-source category leader: Apache-2.0, bring-your-own-key, available as an editor extension, CLI or SDK, with first-class MCP support; you own the keys, the model choice and the (unbounded, usage-based) bill.
Best forDevelopers who want free, model-agnostic AI pair programming in their terminal
The terminal purist’s choice: free, model-agnostic, and the only one here that commits every AI edit to git automatically, making changes genuinely safe to review and revert.
- #15
OpenCode
Open source AI coding agent for the terminal, desktop, browser and IDE, running any model you connect.
Best forDevelopers who want a genuinely free, fully open-source coding agent that works with any model provider, including their own local Ollama models, without being locked into one vendor's terminal tool.
MIT-licensed and free at the core, connecting to 75+ model providers plus local Ollama, with a tested model marketplace behind it: among the most actively-shipped open agents in the category, with a disclosed (now-patched) January 2026 security history worth reading first.
- #16
OpenHands
Open-source software engineering agent with a browser client, CLI and Python SDK; local, cloud or self-hosted.
Best forOpen source-minded engineering teams and enterprises that want a self-hostable coding agent platform
The enterprise self-hosted pick: MIT-licensed core with SSO/RBAC/audit logging and an SDK teams can embed in their own platform (its enterprise/ directory alone runs a separate, source-available license), but bring your own LLM key and budget real setup time.
New open-source session-tree agents
- #17
Juggler
Open-source desktop workbench for AI coding agents, with inspectable tool calls and branching threads.
Best forDevelopers who want hands-on visibility into what an AI coding agent is doing to their codebase and accept early-stage, self-hosted, open-source software.
A genuinely different idea: your session is an editable Yjs tree instead of a scrolling log, built by JUCE creator Julian Storer: free and open source under AGPLv3, still flagged as alpha software on its own README and best treated as a research project for now.
- #18
zot
Single-binary terminal coding agent harness in Go: five built-in tools, most providers, extensions in any language.
Best forDevelopers who want a fast, free, Go-native coding-agent harness with real session-branching control and are comfortable running fast-moving, self-described "beta forever" software from an independent maintainer.
A single Go binary with the same fork-any-message tree-view idea as Juggler, plus a Telegram bridge for supervising a session from your phone: free and MIT-licensed, though its own author calls the UI "98% vibe coded."
Code review agents
- #19
CodeRabbit
AI code review agent for pull requests, the IDE and the CLI, with pre-merge checks and security scans.
Best forEngineering teams, using a coding agent or writing code by hand, that want automatic, consistent PR review plus triage, explanation and continuous security across GitHub, GitLab, Bitbucket or Azure DevOps before a human reviewer looks at it.
The one listing in this category built to review pull requests rather than write code: a genuine free tier and the broadest code-host coverage (GitHub, GitLab, Bitbucket, Azure DevOps), but API access and self-hosting are Enterprise-only.
Terminal-native agents
- #20
Muse Code
Meta's coding agent for planning, editing and validating repository changes from the terminal and CI.
Best forCost-sensitive teams willing to trade code privacy for Contributor tier's roughly 12x cheaper input and 21x cheaper output, or anyone with a large multi-part refactor that benefits from several sub-agents working in parallel isolated worktrees.
Meta's own terminal-native agent, launched in early August 2026, with a Contributor tier that prices tokens 12–21x cheaper than Standard in exchange for Meta training its models on your code: an explicit priced tradeoff neither Claude Code nor Codex CLI offers, and the only such pricing-mirror option among the terminal CLIs here.
In-editor IDE agents
- #21
Junie
JetBrains' coding agent: runs in JetBrains IDEs, the terminal and CI, on JetBrains models or your own keys.
Best forDevelopers and teams already standardized on a JetBrains IDE who want an agent that shares the same debugger, project index, and version control already built into their editor.
JetBrains' own coding agent, deepest inside the IDEs it already owns: Agentic Debugging drives the real IDE debugger, and Plan Mode reviews a change before any code is written. Priced in shared AI credits, not a flat per-seat fee.
Fully autonomous cloud agents
- #22
Augment Code
Cosmos platform running fleets of coding agents across the SDLC, plus Auggie CLI and IDE extensions.
Best forEngineering leaders at organizations who want AI agents running full SDLC loops end to end, with humans at the quality gates, rather than inline completion assistants.
Built around Cosmos, an orchestration layer that runs full loops (code review, ticket-to-PR, security remediation, incident response), rather than one-shot suggestions, with agents runnable in Augment's managed cloud or a team's own AWS/GCP account. Published pricing ($100/month Business, up to 50 seats) is recent and concrete, but every performance number remains vendor-reported.
In-editor IDE agents
- #23
Gemini Code Assist
Google's per-seat AI coding assistant and agent for IDEs, the terminal and GitHub pull requests.
Best forTeams standardized on VS Code or JetBrains who are already on Google Cloud
Google's own IDE assistant, running Gemini 3 with a 1M-token context window across VS Code, the full JetBrains family, Cloud Workstations and Cloud Shell Editor, but the interesting part, Agent Mode's multi-file editing, is Enterprise-only and still preview, and there's no longer a free tier to try any of it.
Visit Gemini Code Assist ↗Read the review →Get a quote →from $19/user/mo billed annually, or $22.80/user/mo billed monthly Open-source & BYOK agents
- #24
goose
Open source AI agent for coding and general tasks; runs locally as a desktop app, CLI, or ACP server.
Best forDevs who want an open source coding agent that reuses an existing Claude or ChatGPT plan
Apache-2.0 and now governed by the Linux Foundation's Agentic AI Foundation, with a genuine point of difference from the rest of this group: ACP support lets it reuse an existing Claude, ChatGPT or Gemini subscription instead of a separate API bill, and drive the same agent from Zed, JetBrains or VS Code. 70+ MCP extensions and recipes/subagents push it past a simple chat loop, but there's no managed hosting and provider rate limits currently surface as hard failures rather than automatic retries.
- #25
Kilo Code
Open source AI coding agent for VS Code, JetBrains, CLI and cloud, with 500+ models at provider rates.
MIT-licensed and free to self-host, with a hosted Gateway reaching 500+ models at provider rates as an alternative to pure BYOK: pricing runs on two independent axes (a Free Individual / $15/seat/month Teams platform plan, plus Kilo Pass AI-inference credits from $19/month Starter to $199/month Expert, with a 5% processing fee on credit purchases); independent cost reviews report real-world inference spend running well past the $19/month entry price.
In this category, not yet ranked
Best forDevelopers and researchers who specifically need an agent whose memory, skills and identity genuinely persist and evolve across long-running sessions, rather than reset on every call.
Side by side
Scroll across to compare every product. Feature names stay in view.
| Feature | Cursor | GitHub Copilot | Devin Desktop | Zed | Claude Code | Codex CLI | Google Antigravity CLI | Warp | Devin | Replit Agent | Factory | Bolt | Cline | Aider | OpenCode | OpenHands | Juggler | zot | CodeRabbit | Muse Code | Junie | Augment Code | Gemini Code Assist | goose | Kilo Code | Letta |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Description | Coding agent in a desktop editor, terminal CLI, cloud VMs and CI, priced per seat plus token usage. | GitHub's AI coding assistant and cloud agent, from editor completions to pull-request review. | AI IDE with a Kanban command center for running local and cloud coding agents; the renamed Windsurf. | Rust code editor with an Agent Panel running Zed's own agent, ACP external agents, or terminal agent CLIs. | Agentic coding tool from Anthropic that edits your codebase, runs commands, and opens pull requests. | OpenAI's terminal coding agent for inspecting, editing and running repository code from the shell. | Terminal agent from Google Antigravity: an agy TUI plus headless mode for macOS, Linux and Windows. | Agentic development environment: a terminal, a standalone agent CLI, and unattended cloud coding agents. | Cloud AI software engineer that takes a ticket, works in its own VM, and opens a pull request. | AI agent that writes, tests and publishes full apps from a chat prompt in Replit's browser workspace. | Droid coding agents that run in the desktop app, CLI, IDEs, CI and Factory-managed cloud computers. | Browser-based AI agent that builds, hosts and publishes web and mobile apps from prompts. | Apache 2.0 coding agent for IDE, terminal and SDK, running on any model provider you choose. | Open-source terminal AI pair programmer that edits and commits code in your local git repo. | Open source AI coding agent for the terminal, desktop, browser and IDE, running any model you connect. | Open-source software engineering agent with a browser client, CLI and Python SDK; local, cloud or self-hosted. | Open-source desktop workbench for AI coding agents, with inspectable tool calls and branching threads. | Single-binary terminal coding agent harness in Go: five built-in tools, most providers, extensions in any language. | AI code review agent for pull requests, the IDE and the CLI, with pre-merge checks and security scans. | Meta's coding agent for planning, editing and validating repository changes from the terminal and CI. | JetBrains' coding agent: runs in JetBrains IDEs, the terminal and CI, on JetBrains models or your own keys. | Cosmos platform running fleets of coding agents across the SDLC, plus Auggie CLI and IDE extensions. | Google's per-seat AI coding assistant and agent for IDEs, the terminal and GitHub pull requests. | Open source AI agent for coding and general tasks; runs locally as a desktop app, CLI, or ACP server. | Open source AI coding agent for VS Code, JetBrains, CLI and cloud, with 500+ models at provider rates. | Stateful agent harness, cloud and SDK for agents with persistent, git-versioned memory. |
| Category | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Coding agents | Agent frameworks, Agent platforms, Coding agents |
| Pricing | Free | Free | Free | $0 / forever | Free | Free | Free | Free | Free | Free | $20 / month | $0 | Free | Free (Apache 2.0) / forever | Free | Free (Free) | Free | Free | $0 / month | $1.25/M input, $4.25/M output / per token | Free | $100/month | $19/user/mo billed annually, or $22.80/user/mo billed monthly | Free (self-hosted) | Free | $0 / month |
| API | Yes | Yes | Yes | No | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | Yes | Yes | Yes |
| Tags |
Get quotes from the top agents
Describe what you need. We route it to the best-matched providers.
Frequently asked questions
- What is the best AI coding agent?
- It depends on where you want to work. For the smoothest in-editor GUI experience, Cursor. For the safest enterprise default with the widest IDE coverage, GitHub Copilot. For a scriptable terminal agent that fits CI, Claude Code. For a fully autonomous agent you delegate a whole ticket to, Devin — or a full-SDLC platform coordinating specialist agents across triage, testing and release, Factory. For a zero-setup browser app builder, Bolt.new or Replit Agent. For a free, open-source tool you fully control, Cline, Aider, OpenCode, the self-hostable, enterprise-ready OpenHands, or Goose if you'd rather reuse an existing Claude, ChatGPT or Gemini subscription than manage a separate API key. For running Claude Code and Codex side by side in one AI-native terminal instead of picking just one, Warp. And if you want an agent that reviews the code any of the above produce rather than writes it, CodeRabbit.
- Which AI coding agents are free or open-source?
- Cline, Aider, OpenCode, OpenHands, Kilo Code, Juggler, Zot, Warp, Zed and Goose are all free, open-source tools (Warp's terminal client went dual AGPL-3.0/MIT in April 2026 — Oz, its cloud-agent orchestration platform, remains closed-source; Zed's core is GPL-3.0-or-later plus Apache-2.0 for select components; Goose is Apache-2.0 and governed by the Linux Foundation's Agentic AI Foundation) — you pay only your chosen model provider (or, for Warp/Zed, work within a metered credit/prediction allotment on their managed plans) (OpenHands' enterprise/ directory carries a separate, source-available license, but its self-hostable core is MIT; Goose is Rust-built and MCP-native, with no enterprise-gated fork). Cursor, GitHub Copilot, Devin, Replit Agent, Windsurf (Devin Desktop), Bolt.new, CodeRabbit and Antigravity CLI all offer a genuine free tier too, though the underlying tool stays closed-source and paid tiers add real usage limits — Factory is the one commercial listing here with no free tier or trial at any price point.
- What is the difference between an in-editor agent and a fully autonomous one?
- In-editor agents like Cursor, GitHub Copilot and Windsurf work interactively while you watch, proposing diffs you review as you go. Devin and Replit Agent instead run asynchronously on their own machine — you hand off a task and come back to a finished pull request or a deployed app rather than pairing with the agent live. Factory takes this furthest: instead of one agent, it coordinates a team of specialized "Droid" agents (code, review, test, docs, monitor) across an entire engineering pipeline, from triage through production monitoring.
- Which coding agent is cheapest to try?
- Cline, Aider, OpenCode, OpenHands, Kilo Code, Juggler, Zot and Goose cost nothing beyond your own model API key (or, for Goose, an existing Claude/ChatGPT/Gemini subscription via ACP), and Google's Antigravity CLI has a genuine $0/month tier too (with Gemini and Claude model access included, on a weekly rate limit). Among commercial tools, GitHub Copilot's Free tier (2,000 completions/month), CodeRabbit's Free tier (unlimited repos for automated PR review) and Claude Code's $17/month Pro plan are the lowest-cost entry points; OpenAI Codex CLI is bundled into ChatGPT's own Free plan too, though regular use typically runs $100–200/month per OpenAI's own figures.
- Is there a fully open-source alternative to Cursor or GitHub Copilot?
- Yes — Kilo Code (an MIT-licensed VS Code/JetBrains/CLI/Cloud extension with a hosted Gateway reaching 500+ models at provider rates), Cline (a VS Code/JetBrains extension), OpenCode (terminal, desktop and IDE), OpenHands (a self-hostable agent platform with enterprise SSO/RBAC on top of its MIT-licensed core), Goose (a Rust-built CLI and desktop app that also plugs into Zed, JetBrains or VS Code via the Agent Client Protocol) and Zed (a from-scratch native editor, not a VS Code fork) are all free, open-source and let you bring any model provider, including local Ollama models. Zed is the only one of the six that's a full standalone editor rather than an extension, desktop app or terminal tool, and it runs Claude Code, Codex or other agents natively via the Agent Client Protocol it created — the same protocol Goose itself supports. The tradeoff versus Cursor or Copilot is that you manage your own API keys (or, for Goose, an existing model subscription via ACP), hosting and costs directly instead of paying one flat subscription.
- Is there an AI agent that reviews code instead of writing it?
- Yes — CodeRabbit automatically reviews pull requests on GitHub, GitLab, Bitbucket and Azure DevOps (plus in-IDE review in VS Code, Cursor and Windsurf) instead of generating new code, so it pairs with any coding agent in this list rather than replacing one. Its Free tier covers unlimited public and private repos, with paid per-seat tiers adding linters, SAST tools and an agentic review chat.
Advertise here
Reach buyers mid-decision. Reach builders choosing their next agent. Promote your brand with a display placement or bring your listing into focus with Featured.
Explore owner options →Advertise on this page →The digestFree
Which agents actually ship.
What we re-checked, what got added, and one number from the index. Tuesdays.
One-click unsubscribe