vs

Claude Code vs ChatGPT: Which should you use for coding?

Facts checked Oct 22, 2026. Every number links to its source.

Summary

Derived from sourced facts, not from builder opinions

ChatGPT includes Codex, OpenAI's coding agent, on every plan from Free up. Claude Code is not in Claude Free. It starts with Pro at $20 a month, or with a pay-per-token Anthropic Console account.
No test covers ChatGPT's regular chat as a coding tool. On the official Terminal-Bench 4.0 board, Claude Code with Opus 5.5 scored 64.85% and Codex, the coding agent that comes with ChatGPT plans, scored 58.18% with GPT-6 Astra or GPT-6.1 Sol.
GPT-6.1 Sol, OpenAI's model for long coding and agent work, runs in Codex and ChatGPT Work, not in regular chat.
Claude Code is not in Claude's Free plan. It starts with Pro at $20 a month, or $17 billed annually.

Choose Claude Code if

You want the agent with the top Terminal-Bench 4.0 result (Claude Code with Opus 5.5).
You already pay for Claude Pro or Max.
You use Claude through Bedrock, Google Cloud or Foundry.

Choose ChatGPT if

You want OpenAI's GPT-6 models for coding.
You want to start free or at $8 a month.
You want published usage numbers per model for agent work (Codex and ChatGPT Work).

Claude Code

ChatGPT

Follow both to get an email when either one changes its price, limits or behavior.

Best by task

Coding agent: Claude Code against Codex, which ChatGPT plans include (Terminal-Bench 4.0 official board)

Claude Code

64.85% for Claude Code with Opus 5.5 against 58.18% for Codex with GPT-6 Astra or GPT-6.1 Sol, all at max effort. ChatGPT's regular chat has no entry.

A coding agent on a free plan

ChatGPT

ChatGPT Free includes Codex with GPT-6 Luna in the desktop app, still rolling out. Claude Code is not in Claude's Free plan.

Published usage limits

ChatGPT

Codex publishes local messages per 5 hours per model on Plus, shared with ChatGPT Work. These do not cover regular chat. Claude Code publishes 5-hour and weekly windows without per-plan numbers.

Plans

Claude Code plans

Pro: $20/month
Max: $100/month
Team Standard seat: $25/month
Team Premium seat: $125/month
Enterprise: $20/month

ChatGPT plans

Free: Free
Go: $8/month
Plus: $20/month
Pro: $100/month
Business: $25/month

Subscription agents are priced per seat with usage limits, so this compares plans rather than dividing a subscription by a guessed number of tasks.

Side by side

	Claude Code	ChatGPT
Maker	Anthropic	OpenAI
Platforms	Terminal CLI, VS Code, JetBrains, desktop app (macOS, Windows, Linux beta), web, iOS and Android app	Not published
Plans	Not in Free. Pro $20/month ($17 annual) · Max from $100 (5x or 20x Pro usage) · Team Standard $25/seat monthly ($20 annual), Premium $125 ($100 annual) · Enterprise $20/seat plus usage at API rates	Free $0 · Go $8 · Plus $20 · Pro $100, $200 or $500 · Business $25/user monthly ($20 annual, 2+ users) · Enterprise and Edu by sales
Usage limits	Per-seat 5-hour and weekly windows · weekly Claude Code limits 25% above pre-May 2026 levels since 2026-09-14 · no published per-plan numbers	Not published
Models	Built for Claude models: aliases opus (Opus 5.5), sonnet (Sonnet 5.5), haiku (Haiku 5.5), fable (Fable 5.1, may bill usage credits) · third-party providers Bedrock, Google Cloud, Foundry	Not published
Use with other tools	Subscription OAuth may not be used to route requests for third-party products	Not published
Models in the app	Not published	GPT-6 Sol and GPT-6 Luna replaced GPT-5.6 in ChatGPT on 2026-10-07 for free and paid plans · Codex and ChatGPT Work still run the September versions of GPT-6 Sol and Luna
GPT-6 Astra access	Not published	Plus, Pro, Business and Enterprise plans
Agent and coding	Not published	ChatGPT Work and Codex are included in every plan, on Free and Go with GPT-6 Luna in the desktop app, subject to rollout
Work and Codex usage limits	Not published	Published only for ChatGPT Work and Codex, which share one allowance, not for regular chat: on Plus about 15-160 local messages per 5 hours with GPT-6.1 Sol and 5-45 with GPT-6 Astra · Pro has no five-hour limit · weekly limits may apply

Scores appear side by side only when both were published for the same test. Settings differ between sources, so small gaps can reverse under another harness.

Updates

Claude Code

27 recorded incidents in the last 90 days.

Default changeOct 7, 2026

Claude Code makes Haiku 5.5 the default Haiku model

Claude Code 2.1.293 added Claude Haiku 5.5 as the default Haiku model on the Anthropic API.

BenchmarkOct 6, 2026

Opus 5.5 in Claude Code tops the official Terminal-Bench 4.0 leaderboard at 64.8%

Claude Code with Opus 5.5 at max effort resolved 64.85% of Terminal-Bench 4.0 tasks (plus or minus 3.1, 330 trials, $4,718.65 total cost), ranked first. The entry was posted October 6, 2026.

BenchmarkOct 6, 2026

Sonnet 5.5 in Claude Code scores 61.8% on Terminal-Bench 4.0

ChatGPT

19 recorded incidents in the last 90 days.

Default changeOct 28, 2026

GPT-6 Luna in ChatGPT updated on October 28 with more concise replies

OpenAI updated GPT-6 Luna in ChatGPT, which launched there on October 7. The updated model gives more concise responses and reduces overly agreeable language.

After

GPT-6 Luna (October 28)

Plan changeOct 28, 2026

ChatGPT Pro and Business add up to $500 a month in free API credits

Eligible ChatGPT Pro and Business subscribers can claim up to $500 per month in free API credits as a subscription benefit. The benefit is activated from ChatGPT or the API Platform by linking an API organization you own.

After

Up to $500 per month in free API credits

Who is affected:: ChatGPT Pro and Business subscribers who also use the API.
What to do:: Link an API organization you own in ChatGPT or the API Platform to claim the credits.

SuccessorOct 15, 2026

GPT-Live-1.1 and GPT-Live-1.1 Quip replace GPT-Live-1 as ChatGPT voice models

GPT-Live-1.1 succeeds GPT-Live-1 and GPT-Live-1.1 Quip succeeds GPT-Live-1 mini. GPT-Live-1.1 is the default voice model for Plus and higher plans, including in Codex and dots, and GPT-Live-1.1 Quip is the default for everyone else.

Before

GPT-Live-1 and GPT-Live-1 mini

After

GPT-Live-1.1 and GPT-Live-1.1 Quip

FAQ

Is Claude Code better than ChatGPT for coding?

No test covers ChatGPT's regular chat as a coding tool. On the official Terminal-Bench 4.0 board, Claude Code with Opus 5.5 scored 64.85% and Codex, the coding agent in ChatGPT plans, scored 58.18%.

Is Claude Code free?

No. Claude Code is not in Claude's Free plan. It starts with Pro at $20 a month, or $17 a month billed annually.

Does ChatGPT have a coding agent?

Yes. Codex comes with every ChatGPT plan. Free and Go get GPT-6 Luna in the desktop app, still rolling out. Plus and higher plans add Codex on the web, in the CLI, in IDEs and on iOS.

Can Claude Code use GPT models?

Not as a built-in choice. Claude Code's model settings list Claude models through the opus, sonnet, haiku and fable aliases. Other makers' models need an API that accepts Anthropic's format, set up and billed separately.

Can ChatGPT edit files on my computer?

Its Codex agent can, in the desktop app, the CLI and the IDE extension.

How much does Claude Code cost?

It comes with Claude Pro at $20 a month ($17 billed annually) and Max from $100.

Does Claude Code publish its usage limits?

No. Each seat has 5-hour and weekly windows, with no published message counts.

Related comparisons

OpenAI Codex vs Claude Code Cursor vs Claude Code Claude vs ChatGPT Gemini app vs ChatGPT OpenClaw vs Claude Code Claude Code vs GitHub Copilot Google Antigravity vs Claude Code ChatGPT vs Grok ChatGPT vs DeepSeek Meta AI vs ChatGPT Microsoft Copilot vs ChatGPT ChatGPT vs Perplexity Gemini CLI vs Claude Code Claude Code page ChatGPT page

Sources and method

Prices, specs and limits come from each maker's own pages. Benchmark scores come from independent benchmark publishers and the makers' announcements, with the setting each was measured in. Changes come from changelogs, deprecation pages and status pages. The summary is written from these facts only.