Vital signs
Why it's on the table
On the table, Coding Agent (Ca) is seat 5 of 58, in the Build family. It is a stable element — the category has settled, choosing is cheap, and switching is rare. Pick a holder and move on; this is not where your decision budget should go. It is marked essential: a company that leaves this seat empty is running with a gap that compounds. It sits in the mid price band — a real line item that should earn its keep visibly.
Coding Agent: the top 5 — v2026.Q3
Edition v2026.Q3 · ranking, pricing and status verified 2026-08-06.
1Claude CodeAnthropic
Pro $17–20/mo · Max $100–200/mo · Team from $20/seat · Enterprise seat + API usageBest for Deep, multi-file agentic work — architecture, refactors, ship-a-feature-from-a-sentence — from terminal, IDE, CI, or phone.
Leads every independent satisfaction metric: 46% "most loved" vs Cursor's 19% and Copilot's 9%, 91% CSAT, NPS 54 (JetBrains AI Pulse, Jan 2026); 44% of developers choose it specifically for complex tasks; #1 on the Terminal-Bench 2.1 harness leaderboard (83.8%, Jun 2026). $2.5B annualized run-rate by Feb 2026 — the fastest-scaling dev tool ever.
Watch Cost at heavy use — power users outgrow Pro fast and land on Max tiers. Copilot still out-adopts it inside 10,000+-employee orgs.
2CursorAnysphere → SpaceX (closing Q3 2026)
Hobby free · Pro $20/mo · Pro+ $60 · Ultra $200 · Teams $40/userBest for Founders who live in an IDE and want the fastest inner loop — tab completion, inline agent edits, and background cloud agents in one polished editor.
The fastest-growing software company on record: $1B ARR (Nov 2025) → $3B (May 2026), 1M+ paying users, 50,000 enterprise customers — and the subject of the largest startup acquisition ever, SpaceX's $60B all-stock purchase agreed June 16, 2026.
Watch Post-acquisition model-routing risk: Cursor depends on Anthropic/Google models and the new owner has its own (Grok). Pricing and limits have changed repeatedly; large complex changes still favor Claude Code/Codex.
3OpenAI CodexOpenAI
Bundled with ChatGPT: Plus $20/mo · Pro $100–200/mo · Business $25/user; token credits since Apr 2026Best for Maximum autonomous task volume per dollar — parallel cloud agents, background automation, and near-benchmark-top models on a $20 entry.
The volume play: user base exploded from under 1M weekly (Feb 2026) to ~9M (Jul 2026) after the GPT-5.6 launch; business usage grew 6x Jan–Apr 2026; sits within a point of the harness-leaderboard top (Codex + GPT-5.5, 83.1% Terminal-Bench 2.1).
Watch Capacity strain moves limits under your feet — context was cut from 372k to 272k tokens during the July surge. OpenAI models only; the IDE-native experience trails Cursor.
4GitHub CopilotMicrosoft / GitHub
Free tier · Pro $10/mo · Pro+ $39 · Business $19/user · Enterprise $39/user; usage-based AI credits since Jun 2026Best for Teams standardized on GitHub — cheapest paid entry, deepest PR/CI integration, easiest org-wide rollout, 90% of the Fortune 100.
The installed-base king: 4.7M paid subscribers (+75% YoY, Microsoft FY26 Q2), ~42% paid share, 56% adoption in 10,000+-employee orgs. Multi-model (OpenAI, Anthropic, Google) with enterprise policy controls nobody else matches.
Watch Lowest satisfaction of the leaders — 9% "most loved", and 29% workplace usage against 42% paid share implies shelfware seats. The June 2026 usage-based billing shift produced 10–50x bill spikes for agentic power users.
5OpenCodeSST (open source, MIT)
Free, MIT-licensed · bring any API key or existing Claude/ChatGPT/Copilot subscription · optional pay-per-use Zen gatewayBest for Lock-in-averse and cost-sensitive builders who want one agent harness over 75+ model providers — including open-weight and local models.
The definitive open-source agent: ~190k GitHub stars, ~900 contributors, top write-in of the State of AI 2026 survey (389 mentions). As near-frontier open models close the benchmark gap at a sixth of the cost, a provider-agnostic harness captures that arbitrage in a way no closed subscription can.
Watch No vendor support — community only, and quality tracks whatever model you plug in. Unmanaged token spend on premium models can exceed a flat $20 plan.
Coding Agent: the top 8 compared
Edition v2026.Q3 · ranking, pricing and status verified 2026-08-06.
| Tool | Pricing model | Free tier | Open source | Runs in | Autonomy | MCP | Enterprise | Models |
|---|---|---|---|---|---|---|---|---|
| Claude Code | Sub + usage | No (trial) | No | Terminal · IDE · web · mobile · CI | Agentic + cloud agents | Yes (native) | SSO · policy · spend | Anthropic |
| Cursor | Sub + usage | Yes | No | IDE · CLI · cloud · CI review | Autocomplete → agentic | Yes | SSO · analytics | Multi-model |
| OpenAI Codex | ChatGPT sub + credits | Limited | CLI only | CLI · IDE ext · web · cloud | Agentic + parallel | Yes | Via ChatGPT Ent. | OpenAI only |
| GitHub Copilot | Sub + credits | Yes | No | Every IDE · CLI · web · CI | Autocomplete → agentic | Yes (allowlists) | Strongest org policy | Multi-model |
| OpenCode | Free + your tokens | Yes (tool) | Yes (MIT) | Terminal · desktop · IDE · CI | Agentic + subagents | Yes | None (community) | 75+ providers |
| Devin Desktop | Sub + credits | Yes | No | IDE · CLI · cloud | Agentic + long-running | Yes | Yes | Multi (SWE-1.5+) |
| Aider | Free + your tokens | Yes (tool) | Yes (Apache) | Terminal | Pair-programming | Limited | No | Any (BYOK) |
| Cline | Free + your tokens | Yes (tool) | Yes (Apache) | VS Code · JetBrains · CLI | Agentic, approval-gated | Yes + marketplace | Teams tier | Any (BYOK) |
How to choose your coding agent
- If your bottleneck is hard engineering — architecture, gnarly refactors, multi-repo work
- Claude Code. 44% of developers pick it specifically for complex tasks; pair heavy use with the Max tier rather than fighting Pro limits.
- If you live in an editor and the inner loop is everything
- Cursor — accepting the post-SpaceX model-routing uncertainty as the price of the polish.
- If you want maximum autonomous throughput per dollar, or background agents running in parallel
- Codex on ChatGPT Plus at $20 — the cheapest seat at the benchmark top table.
- If procurement, compliance, or a GitHub-standard org decides for you
- Copilot Business — the 90%-of-Fortune-100 path, with the honest caveat that its own users love it least.
- If you refuse lock-in, run high token volume, or want open-weight models
- OpenCode — free harness, any provider, and the open-model price-performance keeps improving quarterly.
Coding Agent: the whole field
30 more tools tracked in this category, including 5 dead, renamed, or sunsetting — a reference that hides the graveyard isn't one. Verified 2026-08-06.
| Tool | Maker | What it is | Entry | Status |
|---|---|---|---|---|
| Devin / Devin Desktop | Cognition | Autonomous cloud engineer + the ex-Windsurf IDE (rebranded Jun 2026); $492M ARR, $26B valuation — the near-miss for the top 5 mid-rebrand | $20/mo + credits | active |
| Amp | Amp, Inc. (ex-Sourcegraph) | Frontier agentic CLI, spun out independent Dec 2025; novel ad-supported free tier | usage-based | active |
| Aider | community (Apache-2.0) | The original git-native terminal pair programmer; still excellent for structured refactors, momentum slowing | free + tokens | active |
| Cline | Cline Bot Inc. | Leading open-source agentic VS Code/JetBrains extension (~64k stars) + CLI and SDK | free + tokens | active |
| Kilo Code | Kilo | Cline+Roo hybrid fork with Architect/Code/Debug modes | free + tokens | active |
| Zed | Zed Industries | GPU-fast Rust editor with agentic editing and ACP support | free · Pro ~$20/mo | active |
| JetBrains AI / Junie | JetBrains | AI Assistant + Junie agent across the IntelliJ family; Junie CLI GA Jun 2026 | free · Pro $10/mo | active |
| Antigravity | Agent-first IDE (Gemini 3), free public preview; Google's flagship after retiring Gemini CLI for individuals | free preview | active | |
| Jules | Async cloud coding agent, GitHub-connected | free tier | active | |
| Kiro | AWS | Spec-driven agentic IDE + CLI — now AWS's flagship, replacing Amazon Q Developer (EOL Apr 2027) | free · Pro $20/mo | active |
| Factory (Droid) | Factory AI | Enterprise agent platform (CLI/IDE/Slack); $150M Series C at $1.5B (Apr 2026) | usage-based | active |
| Warp | Warp | Agentic development environment terminal; core open-sourced 2026 | free · Build $20/mo | active |
| Trae | ByteDance | Free-heavy AI IDE with SOLO autonomous mode | free · Pro ~$10/mo | active |
| Qodo | Qodo (ex-Codium) | Code-quality agents for review and testing — overlaps element Cq | free tier | active |
| Tabnine | Tabnine | Privacy-first, air-gap-deployable enterprise assistant; Gartner MQ Visionary (May 2026) | ~$39/user/mo | active |
| Continue | Continue Dev | Open-source IDE extension + CI-first CLI for headless PR checks | free + tokens | active |
| Goose | Block → AAIF foundation | MCP-native local agent framework; ran to 60% of Block internally; foundation-governed since Dec 2025 | free + tokens | active |
| OpenHands | All Hands AI | Open agent platform (ex-OpenDevin, ~79k stars) with event-sourced replay and CI SDK | free + tokens | active |
| Crush | Charmbracelet | Polished TUI terminal agent in the OpenCode lineage, LSP + MCP | free + tokens | active |
| Mistral Vibe | Mistral AI | Open-source CLI agent + Devstral 2 open-weight coding model | free + tokens | active |
| Kimi Code CLI | Moonshot AI | MIT-licensed CLI with coder/explore/plan subagents (Jun 2026) | free | active |
| Grok Build | xAI | CLI agent with worktree-isolated parallel subagents (May 2026) | unverified | active |
| Qwen Code / Qoder | Alibaba | OSS CLI (BYOK; free hosted tier ended Apr 2026) + Qoder agentic IDE | free + tokens | active |
| CodeBuddy | Tencent | IDE + CLI agent on DeepSeek/Hunyuan models | free tier | active |
| Pi / Omp | community (MIT) | Minimal-core CLI agent wave — 4 tools, sub-1k-token prompt, ~67k stars | free + tokens | active |
| Open Interpreter | Open Interpreter | Pioneer local code-execution agent; development quiet through 2026 | free | fading |
| Gemini CLI | Free individual access ended Jun 18, 2026 — enterprise-only via Code Assist; succeeded by Antigravity CLI | enterprise only | sunsetting | |
| Roo Code | Roo Code Inc. | Shut down May 15, 2026; team pivoted to Roomote — successors: Zoo Code, Kilo Code | — | dead |
| Amazon Q Developer | AWS | Sunsetting into Kiro; signups closed May 2026, end-of-life Apr 30, 2027 | — | sunsetting |
| Windsurf | Cognition | Brand retired Jun 2, 2026 — became Devin Desktop after the 2025 Google licensing / Cognition acquisition saga | — | renamed |
Coding Agent: the category in numbers
Edition v2026.Q3 · ranking, pricing and status verified 2026-08-06.
- AI code-tools market ~$9.35B (2026) → $29.96B by 2031, 26.2% CAGR [src]
- Cursor: $1B ARR (Nov 2025) → $3B (May 2026); SpaceX $60B acquisition agreed Jun 16, 2026 — the largest startup purchase ever [src]
- Claude Code: $1B (Nov 2025) → $2.5B annualized (Feb 2026); majority enterprise [src]
- Cognition (Devin + ex-Windsurf): $1B raised at $26B post (May 2026); $492M ARR [src]
- Trust gap persists: AI-generated code stalls at a 56% security pass rate (Veracode, Jul 2026); 46% of devs distrust accuracy (Stack Overflow 2025) [src]
- 2026 structural shift: free tiers retreating (Gemini CLI, Qwen hosted, Amp waitlist), usage-based billing spreading (Copilot, Kiro, Replit), first OSS casualties (Roo Code) [src]
Coding Agent: method & sources
Benchmark caveat: model-level leaderboards (e.g. llm-stats) and harness-level leaderboards (tbench.ai, agent+model pairs) disagree on the leader; we cite the harness leaderboard because it measures the product you actually install. Survey caveat: Stack Overflow's 2026 survey opened June 23 and has no published results yet — anything citing '2026 Stack Overflow results' is recycling 2025 data. App-builders (Lovable, Replit, Bolt, v0) belong to element Vb · Vibe Builder, not here. Ranking criteria: verified commercial traction, independent satisfaction surveys, agent benchmarks, and founder-fit (price floor, lock-in, surfaces). Editorial, never paid — the charter. Machine-readable twin: ca.json.
All sources (19)
- https://www.tbench.ai/leaderboard/terminal-bench/2.1
- https://blog.jetbrains.com/research/2026/04/which-ai-coding-tools-do-developers-actually-use-at-work/
- https://2026.stateofai.dev/en-US/coding-agents-assistants/
- https://sacra.com/c/anthropic/
- https://en.wikipedia.org/wiki/Cursor_(company)
- https://techcrunch.com/2026/06/16/spacex-to-acquire-cursor-for-60b-in-stock-days-after-blockbuster-ipo/
- https://techcrunch.com/2026/05/27/ai-coding-startup-cognition-raises-1b-at-25b-pre-money-valuation/
- https://axis-intelligence.com/github-copilot-statistics/
- https://thenewstack.io/gpt-5-6-codex-user-surge/
- https://github.blog/news-insights/company-news/github-copilot-is-moving-to-usage-based-billing/
- https://www.veracode.com/blog/2026-genai-code-security-report-ai-risk/
- https://blog.arcbjorn.com/state-of-cli-coding-agents-2026
- https://www.mordorintelligence.com/industry-reports/artificial-intelligence-code-tools-market
- https://survey.stackoverflow.co/2025/ai/
- https://opencode.ai/docs
- https://claude.com/pricing
- https://cursor.com/pricing
- https://github.com/features/copilot/plans
- https://developers.openai.com/codex
Our take
The single highest-leverage element on the table. If you hold only three elements, this is one of them.
Combines with
Appears in compounds
Guides
This is element 5 of 58. The table is versioned quarterly — when a tool loses its seat, the changelog records the succession.
Explore the full table →