know.2nth.ai Agents Coding CLI agents
agents · coding CLI comparison

Claude Code·Codex CLI·Copilot CLI

The three agentic coding CLIs that matter in mid-2026 — Anthropic's Claude Code, OpenAI's Codex CLI, and GitHub Copilot CLI. All three sit in the terminal, read codebases, plan multi-file changes, run tests, and iterate on failures. They differ in model lock-in, sandbox model, pricing shape, and extensibility. This leaf maps where each wins.

Reference comparison Agents domain Hot · quarterly review

Three terminal-native coding agents. Different design centres.

All three operate at the project level rather than line-by-line autocomplete. Read codebases, plan changes, edit multiple files, run tests, iterate. The differences begin at the model layer.

Claude Code

Anthropic's coding agent

The same engine the Claude Agent SDK runs on. Locked to Claude (Sonnet, Opus). Native Skills, hooks, MCP, sub-agents, plan-mode. Bundled with Claude Pro / Max / Team / Enterprise subscriptions.

Deep dive on Claude Code →

OpenAI Codex CLI

Cloud-sandbox parallel agent

Open-source CLI binary (Rust, MIT) backed by OpenAI's codex-1 (o3-based) and codex-mini (o4-mini-based) models. Cloud-sandboxed by default; parallel task execution. Bundled with ChatGPT Plus / Pro / Business / Enterprise.

Deep dive on Codex CLI →

GitHub Copilot CLI

Multi-model GitHub-native agent

Routes between Claude (Sonnet/Opus 4.6) and GPT-5.2–5.5 from one CLI; switch with /model. Worktree-isolation, self-iterating code review, Plan vs Autopilot modes. Bundled with Copilot Pro / Pro+ / Business / Enterprise; moves to usage-based AI Credits June 1 2026.

Deep dive on Copilot CLI →

Fifteen dimensions, scored honestly.

Where the three differ in shape. "Different" is more common than "better."

DimensionClaude CodeCodex CLICopilot CLI
MakerAnthropicOpenAIMicrosoft / GitHub
CLI licenseProprietaryOpen source (Rust)Proprietary
ModelClaude only (Sonnet, Opus)OpenAI only (codex-1, codex-mini)Multi-model — Claude 4.5/4.6 + GPT-5.2–5.5
Switch modelsn/a (single model)n/a (OpenAI-only)/model mid-task
Where it runsTerminal · VS Code · JetBrains · claude.ai/code webTerminal · web (codex.app) · IDE · Chrome extensionTerminal · VS Code · JetBrains delegation · Copilot Workspace
Sandbox modelLocal execution; approval-gatedCloud sandboxes by default; parallel tasksWorktree or workspace isolation; local
Approval modelApproval-by-default; Plan + AutopilotApproval-by-defaultPlan (approval) or Autopilot (autonomous)
Sub-agentsYes — context-isolated with restricted toolsCode-review agent built inYes — Explore, Task + custom in ~/.copilot/agents/
MCPNative; first-classNativeNative
Skills / customisationSkills, hooks, sub-agents, slash commandsModal Vim editing, custom workflows, MCPCustom agents, /delegate, image analysis
IDE bridgeVS Code + JetBrains extensionsVarious IDE integrationsJetBrains delegation (May 2026 preview); VS Code native
Entry pricingPro $20/moChatGPT Plus $20/moCopilot Pro $10/mo + $10 AI Credits
Pricing shapeTiered credits (Pro / Max / Team)Rolling 5-hr usage windows; tier multipliersUsage-based credits (June 1 2026); 1 credit = $0.01
June 2026 changeSDK credits separate from chat credits (Jun 15)Stable as of May 2026Pivot to usage-based AI Credits (Jun 1)
StandoutStripe migrated 10k lines in 4 days vs 10 weeks manualcodex-1 RL-trained on real PRs; cloud-parallelSelf-iterating code review before opening PRs

Three structural choices the others can't easily change.

Below the surface noise, each tool has made one structural commitment that the others can't replicate without redesigning. These are the differences worth weighting.

1 · Model lock-in vs flexibility

Claude Code commits to Claude. Codex CLI commits to OpenAI. Copilot CLI deliberately doesn't commit — you pick the model per task via /model. For teams hedging on model choice (or wanting open weights through frontier without rewrites), Copilot CLI's design is the only one that fits structurally. For teams that have already picked their lane, Claude Code or Codex CLI win on first-party depth.

2 · Local vs cloud sandbox

Claude Code and Copilot CLI are local-first — the agent runs in your terminal, edits your files, runs your tests. Codex CLI defaults to cloud-sandboxed execution — the agent works on a copy in OpenAI's sandbox while you keep coding locally. This makes parallel work natural: fire off three tasks in three sandboxes, review the PRs when they're done. For teams who want "kick off a migration and forget about it for an hour," Codex CLI's cloud-parallel pattern is genuinely different.

3 · Open-source CLI vs proprietary

The Codex CLI binary itself is open source (Rust, npm-installable, Homebrew). Read the source, fork it, swap binaries. The proprietary part is the model behind the OpenAI API it calls. Claude Code and Copilot CLI are fully proprietary CLI binaries that call proprietary models. For teams that want auditable agent-side code (security-sensitive environments, regulated industries), the Codex CLI's OSS posture is unique among the three.

The honest decision matrix.

Three real tools, three real fits. The right pick is rarely "which is best overall" — it's "which matches the team you have, the model you're already paying for, and the workflow shape that fits your repo."

Pick Claude Code if…

You're on the Claude stack

  • The team already pays for Claude Pro / Max
  • You want the deepest single-codebase reasoning — Stripe / Wiz-style multi-file refactors
  • Anthropic-ecosystem extensibility (Skills, hooks, MCP, sub-agents) matches your customisation needs
  • Single-model is a feature, not a limitation — fewer knobs
  • You value safety-first defaults with mature Plan / Autopilot modes
  • Cross-platform IDE story matters (VS Code + JetBrains + web)
Pick Codex CLI if…

You want parallel cloud work

  • The team already pays for ChatGPT Plus / Pro
  • Cloud-sandboxed parallel work matches your workflow — fire off N tasks, review PRs when they finish
  • You want an open-source CLI binary you can audit and fork (Rust, npm)
  • codex-1's RL-on-PRs training matters for your use case
  • Modal Vim editing in the composer matters
  • Cross-tab browser work via the Chrome extension is in scope
Pick Copilot CLI if…

You're GitHub-native + want flexibility

  • The team lives on GitHub — PRs, issues, Actions, code review
  • Multi-model from one CLI — switch Claude / GPT per task without juggling API keys
  • You want the self-iterating code review pattern before opening PRs
  • Worktree isolation matches your branching workflow
  • You'd rather pay usage-based credits than tier-fixed pricing (or vice versa — model carefully)
  • Image analysis from bug-report screenshots is in scope

It's not always either-or.

Running more than one is normal

All three are subscription-bundled with chat tools the team probably already pays for. It's common to see Claude Code for deep refactors, Codex CLI for cloud-parallel tasks, and Copilot CLI for GitHub-native PR work — all on the same machine, swapped per task. Cost of running multiple: minimal (each is included in its parent subscription). Cost of context-switching between mental models: real but bounded.

Update · Gemini CLI is becoming Antigravity CLI

On 19 May 2026 Google announced that Gemini CLI is being transitioned to Antigravity CLI — part of the broader Google Antigravity 2.0 agent-first development platform.

Timeline. Antigravity CLI is available to all users from 19 May 2026. Gemini CLI and the Gemini Code Assist IDE extensions stop serving free and paid individual users (Google AI Pro / Ultra) on 18 June 2026. Enterprise customers on Gemini Code Assist Standard / Enterprise keep Gemini CLI indefinitely, with continued model updates.

What carries over. Agent Skills, Hooks, Subagents, and Extensions (renamed plugins). Antigravity CLI is Go-based (faster execution), adds asynchronous multi-agent workflows, and shares architecture with the desktop Antigravity 2.0 app. Perfect feature parity is not there at launch.

Google's stated reason: individual workflows have outgrown the 2025-era single-agent CLI; users now need multiple agents collaborating. Google announcement →

Open-source alternatives worth knowing

If subscription-gated CLIs don't fit, the credible OSS alternatives in mid-2026:

Aider — Python, git-native, terminal-first; brings your own model API key (Claude, GPT, Gemini, Ollama).

Cline — MIT-licensed VS Code agent; 1.5M+ installs by April 2026; autonomous multi-file execution.

Gemini CLI → Antigravity CLI — Google's offering, transitioning to Antigravity CLI from 19 May 2026 (see the update note above). Individual users migrate by 18 June 2026; enterprise keeps Gemini CLI.

Continue.dev — open-source IDE plugin; multi-model.

OpenHands, Goose, Roo Code, Qwen Code, Kimi CLI — various OSS options.

Cursor (Cursor 3 released April 2026) — IDE-first with cloud agents, /worktree, parallel Agent Tabs; not CLI-shaped but covers the same workflow.

What lands in SA dev shops.

Cost ceiling matters more in SA than in US-based teams. At ZAR/USD exchange rates and SME margins, the difference between $10/mo and $200/mo per developer is material. Copilot Pro at $10/mo (with $10 AI Credits) is the cheapest entry. Claude Pro and ChatGPT Plus at $20/mo are the next rung. Heavy-usage tiers (Pro+ / Max / Pro200) only make sense once you've proved the workflow pays off.

Local model support matters when bandwidth or POPIA-sensitive data is in scope. None of these three CLIs supports fully-local inference natively — the closest is using them as the framework while routing through a model gateway that hits Ollama / vLLM. For "code never leaves the building" environments, Aider + Ollama is the cleanest path.

GitHub is the dominant SCM in SA dev shops. That alone gives Copilot CLI a default-pick gravity for many teams. The friction of "set up GitHub auth" is zero (already done); the friction of "set up a third subscription" is real.

Talent pool consideration. Engineers in SA in 2026 are most likely to have used Cursor or Copilot before any of the three CLIs. Onboarding cost for Claude Code or Codex CLI is real — not insurmountable, just real. Plan a sprint of pair-programming for whichever tool you pick.

Where this leaf links into the rest of the tree.

Primary sources.

All three move fast — verify versions against the date on this leaf.