> ## Content Index
> Fetch the complete content index at: https://claudefox.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Claude Code vs GitHub Copilot (2026): pricing, Claude models, which to use
- URL: https://claudefox.com/claude-code-vs-github-copilot/
- Published: 2026-11-20T09:13:11.000Z
- Updated: 2026-11-20T09:13:18.000Z
- Description: A sourced comparison of Claude Code and GitHub Copilot in November 2026: Copilot's Free, Pro, Pro+, Max, Business and Enterprise plans, which Claude models each one offers, how AI credits replaced premium requests, the cloud agent and Copilot CLI, costs and our take.
- Author: hupicake
- Tags: Tips

Last checked: 2026-11-20, against GitHub's Copilot plans page and docs and Anthropic's Claude Code docs, pricing pages and Help Center. Prices in US dollars before tax.

Pick Claude Code if Claude is the model you want for agent work: a $20 Claude Pro plan includes Claude Code with Opus 5.5 as the default model. Pick GitHub Copilot if you want inline completions that are unlimited on paid plans, agents that open pull requests and a choice of vendors. Copilot offers Claude on every paid plan, with Opus 5.5 on Pro+, Max, Business and Enterprise only, and bills each Claude request in AI credits at per-token prices that match Anthropic's API list prices for Opus 5.5 and Sonnet 5.5.

## Claude Code vs GitHub Copilot at a glance

Claude Code is Anthropic's Claude-only agent, included in paid Claude plans; Copilot is GitHub's multi-vendor assistant that bills chat, agent and CLI requests in AI credits.

__Cells link the vendor page behind each figure, or our guide that covers it. Prices as of November 20, 2026.__
| Criterion          | Claude Code                                                                                                                                                                                                                                 | GitHub Copilot                                                                                                                                                                                                                                                                                                          |
| ------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Pricing and plans  | [Pro $20 a month ($17 billed annually), Max $100, Ultra $200; Team $20 or $100 per seat billed annually. Claude Code is in paid plans only](https://claude.com/pricing?ref=claudefox.com)                                                   | [Free; Pro $10, Pro+ $39, Max $100 a month; Business $19 and Enterprise $39 per seat, each with included GitHub AI Credits (1 credit = $0.01). Copilot access ($0 with GitHub Enterprise Cloud, $10 standalone) includes no credits](https://docs.github.com/en/copilot/get-started/plans?ref=claudefox.com)            |
| Models available   | [Anthropic only: Opus 5.5 is the default on Pro, Max, Ultra, Team, Enterprise and the API; Sonnet 5.5, Haiku 5.5, and Fable when selected (paid with usage credits on Pro)](https://code.claude.com/docs/en/model-config?ref=claudefox.com) | [Claude, GPT, Gemini, Grok, Kimi and Microsoft models. Pro: Sonnet 5.5, Sonnet 5, Haiku 5.5 and 4.5\. Opus 5.5 and Fable on Pro+, Max, Business and Enterprise](https://docs.github.com/en/copilot/reference/ai-models/supported-models?ref=claudefox.com)                                                              |
| Where it runs      | [Terminal, VS Code, JetBrains, desktop app and web; a Slack integration routes bug reports to Claude Code](https://code.claude.com/docs/en/overview?ref=claudefox.com)                                                                      | [VS Code, Visual Studio, JetBrains, Xcode, Eclipse, GitHub.com, GitHub Mobile, Copilot CLI, the Copilot app](https://docs.github.com/en/copilot/reference/ai-models/supported-models?ref=claudefox.com)                                                                                                                 |
| Agent features     | [Subagents, hooks, MCP, skills, plugins and marketplaces](https://code.claude.com/docs/en/sub-agents?ref=claudefox.com)                                                                                                                     | [Agent mode in IDEs, a cloud agent that opens pull requests, code review, hooks, skills, MCP; Claude and Codex as third-party agents (preview)](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents?ref=claudefox.com)                                                                   |
| Context and limits | [Native 1M-token window on Opus 5.5 and Sonnet 5.5; 5-hour and weekly limits (verified Ultra accounts have no 5-hour limit)](https://claudefox.com/claude-code-usage-limits/)                                                               | [1M-token option for most Claude models (not Opus 4.8 fast mode) in VS Code and Copilot CLI only, using more credits. After the monthly pool, individuals set a dollar budget; for organizations extra use is on by default](https://docs.github.com/en/copilot/reference/ai-models/supported-models?ref=claudefox.com) |
| Privacy and data   | [Consumer plans: training is your choice, 5-year retention if you allow it, 30 days if not; Team, Enterprise and API: no training unless you opt in](https://code.claude.com/docs/en/data-usage?ref=claudefox.com)                          | [Individual plans: GitHub may train on prompts and suggestions unless you opt out; GitHub's FAQ describes this training for individual subscribers only](https://github.com/features/copilot/plans?ref=claudefox.com)                                                                                                   |
| Default safety     | [Auto mode with a safety classifier is the starting mode from v2.1.283 (Anthropic says it does not guarantee safety); Manual mode asks first](https://code.claude.com/docs/en/permission-modes?ref=claudefox.com)                           | [Copilot CLI asks before tools run unless you pass approval flags; optional local or cloud sandbox](https://docs.github.com/en/copilot/concepts/agents/about-copilot-cli?ref=claudefox.com)                                                                                                                             |
| Best fit           | Claude-first developers, long terminal sessions, fleets on Bedrock or Google Cloud                                                                                                                                                          | GitHub-centred teams wanting completions, PR agents and code review across vendors on one bill                                                                                                                                                                                                                          |

## Which Claude models does GitHub Copilot offer, and on which plans?

GitHub lists eleven Claude models for Copilot (Opus 5.5, Opus 4.8 and its fast mode, Fable 5, 5.1 and 5.2, Sonnet 5.5, 5 and 4.6, Haiku 5.5 and 4.5), but which ones you can pick depends on the plan: Opus 5.5 and Fable need Pro+ or higher, while Copilot Pro is limited to Sonnet and Haiku.

- Copilot Free and Student: models only through auto model selection, so you cannot choose a Claude model yourself. Free also caps code completions at 2,000 a month.
- Copilot Pro ($10): Claude Sonnet 5.5, Sonnet 5, Haiku 5.5 and Haiku 4.5\. GitHub's plan table marks Opus 5.5 and Fable as not included on Pro.
- Copilot Pro+ ($39): adds Claude Opus 5.5, Opus 4.8 fast mode (preview) and Fable 5, 5.1 and 5.2.
- Copilot Max ($100): the same Opus 5.5, Opus 4.8 fast mode and Fable models, but GitHub's table does not list full Opus 4.8 or Sonnet 4.6 for Max.
- Copilot Business ($19), Enterprise ($39) and Copilot access: Max's Claude line-up, subject to admin policy. Fable is off by default on these plans because Anthropic retains Fable prompts and outputs by default to run safety classifiers. Enterprises can request a zero-data-retention exemption through the end of 2026 (GitHub says Enterprise Frontier Safeguards would be required after that), and an admin must still enable each Fable model.
- Retired models on annual plans: Sonnet 4.6 (retired September 1, 2026) stays available to individuals on annual Pro and Pro+ plans, and Opus 4.8 (retired November 16, 2026) to individuals on annual plans, which GitHub's plan table lists under Pro+ only. GitHub says legacy annual plans will not get new models.

Availability is not included usage: every Claude request is billed in AI credits per token. Copilot lists Opus 5.5 at $4 per million input tokens, $0.20 cached input, $5 cache write and $20 output, and Sonnet 5.5 at $2 input and $10 output. Those match Anthropic's API list prices; Anthropic also lists a pricier 1-hour cache write that Copilot's table does not show. Choosing the 1M window or a higher reasoning level uses more tokens, so more credits.

## Who should pick Claude Code, and who should pick Copilot?

Claude Code suits long Claude agent sessions at a flat price; Copilot suits work across the whole GitHub workflow, with metered model use.

- Pick Claude Code if you want Opus 5.5 for $20 a month. On Copilot, the cheapest plan with Opus 5.5 is Pro+ at $39.
- Pick Copilot if you want tab completions. Inline suggestions and next edit suggestions are unlimited on paid Copilot plans and do not use AI credits.
- Pick Copilot if your work starts on GitHub: assign issues to the cloud agent, get pull request reviews, or let Agent Merge fix failed workflows and merge conflicts.
- Pick Copilot if you want several vendors in one picker, or an admin-controlled pooled budget for a team.

## How much does each cost?

Our reading of the two billing models, not a measured test: for heavy Claude use, Claude Code is likely to be cheaper, because Copilot meters every Claude token against a fixed credit pool while Claude plans include usage up to a limit. Anthropic says there is no fixed message count for those limits, so no direct dollar match is possible; for a few short Sonnet tasks a month Copilot Pro at $10 can cost less.

Copilot's individual plans include base credits equal to the price plus a flex allotment that GitHub says can change with model economics: Pro has 1,000 base and 1,500 total credits ($15 of model use), Pro+ 3,900 and 7,000 ($70), Max 10,000 and 20,000 ($200). Business and Enterprise give 1,900 and 3,900 credits per user, pooled across the organization, with extra use at $0.01 per credit, allowed by default unless an admin turns it off; Copilot access has no included credits and meters from the first use. Individuals who run out set a dollar budget for extra use, upgrade, or wait for the reset at 00:00 UTC on the 1st; unused credits do not carry over. Code review costs credits plus GitHub Actions minutes; GitHub estimates an approximate average of $0.25, $1.50 or $10 per review at the Lite, Balanced and Max effort levels (typical ranges $0.05 to $1, $0.25 to $5 and $1.50 to $25), excluding Actions minutes; the review model is not disclosed.

- One Opus 5.5 task with 1 million input tokens and 50,000 output tokens, without caching: $4 plus $1, so $5, or 500 AI credits. Pro+ covers about 14 such tasks a month from its full 7,000 credits; Copilot Pro cannot run Opus 5.5 at all.
- The same task on Sonnet 5.5: $2 plus $0.50, so 250 credits, about six a month on Copilot Pro's 1,500 credits.
- Claude Pro ($20) counts that work against its 5-hour and weekly limits, not a dollar meter.
- Heavy users: Claude Max costs $100 for 5x Pro usage plus $100 a month in API credits, and Ultra $200 for 10x Pro's weekly usage plus $200 in API credits. Copilot Max costs $100 for 20,000 credits, or $200 of model use at list prices.
- Teams: Claude Team standard seats cost $20 a month billed annually; Copilot Business costs $19 a seat with 1,900 pooled credits.

## What happened to premium requests and model multipliers?

They are legacy: since June 1, 2026, Copilot bills by tokens, and premium request multipliers apply only to people who kept an existing annual Pro or Pro+ plan.

Those annual subscribers still get 300 (Pro) or 1,500 (Pro+) premium requests a month, with extra requests at $0.04 each. Each prompt costs one request times the model's multiplier: 0.33 for Claude Haiku 4.5, 9 for Claude Sonnet 4.6 and 27 for Claude Opus 4.8, and a Copilot code review costs 13\. When the annual plan ends, GitHub moves the account to Copilot Free unless you switch to a monthly plan.

## How do you switch between Claude Code and Copilot?

Keep shared rules in AGENTS.md or CLAUDE.md, because Copilot reads both, and move skills into a folder both tools scan.

- Instructions: Copilot CLI reads AGENTS.md, CLAUDE.md, .claude/CLAUDE.md and GEMINI.md. On GitHub.com, Copilot uses AGENTS.md files anywhere in the repository, or a single CLAUDE.md in the root. Claude Code reads AGENTS.md only when there is no CLAUDE.md in the working directory or above it.
- Skills: Copilot loads project skills from .github/skills, .claude/skills or .agents/skills, so existing Claude Code skills in .claude/skills work in both.
- Hooks: Copilot uses .github/hooks/\*.json with events such as preToolUse and sessionStart; there is no import of Claude Code hook settings, so rewrite them.
- Models: Copilot CLI can also call your own Anthropic API key through COPILOT\_PROVIDER\_TYPE=anthropic, COPILOT\_PROVIDER\_API\_KEY and COPILOT\_MODEL.
- Use both: Pro+, Max, Business and Enterprise can assign GitHub issues to Anthropic's Claude coding agent (public preview, built on the Claude Agent SDK), paid in Copilot credits plus GitHub Actions minutes. GitHub's plan docs also list Pro; its plans page does not.

```
# Install Copilot CLI next to Claude Code
npm install -g @github/copilot   # needs Node.js 22+
copilot                          # run /login on first launch; Shift+Tab cycles modes, including plan mode
```

## Frequently asked questions

- Can I use Claude models in GitHub Copilot? Yes, on every paid plan, billed in AI credits at per-token prices. Opus 5.5 and Fable need Pro+ or a Max, Business or Enterprise plan; Free and Student reach models only through auto selection.
- Is Copilot cheaper than Claude Code? Copilot Pro is cheaper to start at $10 with unlimited completions, but its 1,500 credits buy about $15 of model use.
- Does Copilot have a CLI like Claude Code? Yes. Copilot CLI runs on Linux, macOS and Windows, with plan mode, a -p flag for scripts, MCP, hooks, skills and custom agents, in every plan.
- Does GitHub train on my Copilot data? On individual plans GitHub may use prompts, suggestions and code snippets for training unless you opt out in settings. GitHub's FAQ describes this training for individual subscribers only and does not mention it for Business and Enterprise.
- Can I use my Claude subscription inside Copilot? Not as a login. Copilot bills Claude through GitHub credits, and Anthropic's terms reserve Claude subscription sign-in for Claude Code and Anthropic's own apps. Copilot CLI does accept your own Anthropic API key in place of GitHub-hosted models.

## Our take

This is opinion. If Claude is the model you want, Claude Code is the better deal in November 2026: Opus 5.5 is the default on a $20 plan, while Copilot needs $39 for Opus 5.5 and then meters each token against about $70 of credit. Copilot is the better product if your work lives on GitHub: unlimited completions, issue-to-pull-request agents and code review in one subscription, with Claude in the model picker. Because Copilot reads CLAUDE.md and .claude/skills, running both needs only one set of rules.

## Related on claudefox

- [Claude Code pricing (November 2026): plans and API costs](https://claudefox.com/claude-code-pricing/)
- [Claude Code usage limits explained: 5-hour and weekly caps](https://claudefox.com/claude-code-usage-limits/)
- [How to install Claude Code on Mac, Windows and Linux](https://claudefox.com/install-claude-code/)
- [Claude Code vs Codex CLI (2026): which coding agent to use](https://claudefox.com/claude-code-vs-codex/)
- [Claude Code vs Cursor (2026): pricing, models, which to use](https://claudefox.com/claude-code-vs-cursor/)
- [Claude API pricing calculator: cost per request, day and month](https://claudefox.com/claude-api-pricing-calculator/)

Source: [GitHub Copilot plans and pricing](https://github.com/features/copilot/plans?ref=claudefox.com), [Plans for GitHub Copilot (GitHub Docs)](https://docs.github.com/en/copilot/get-started/plans?ref=claudefox.com), [Supported AI models in GitHub Copilot (GitHub Docs)](https://docs.github.com/en/copilot/reference/ai-models/supported-models?ref=claudefox.com), [Models and pricing for GitHub Copilot (GitHub Docs)](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing?ref=claudefox.com), [Usage-based billing for individuals (GitHub Docs)](https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/billing?ref=claudefox.com), [Usage-based billing for organizations and enterprises (GitHub Docs)](https://docs.github.com/en/copilot/concepts/billing-and-usage/organizations-and-enterprises/billing?ref=claudefox.com), [Usage limits for GitHub Copilot (GitHub Docs)](https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/usage-limits?ref=claudefox.com), [Requests in GitHub Copilot, legacy (GitHub Docs)](https://docs.github.com/en/copilot/concepts/billing/copilot-requests?ref=claudefox.com), [Model multipliers for annual plans, legacy (GitHub Docs)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/model-multipliers-for-annual-plans?ref=claudefox.com), [What changed with Copilot billing, legacy (GitHub Docs)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/what-changed-with-billing?ref=claudefox.com), [GitHub Copilot on GitHub.com and GitHub Mobile (GitHub Docs)](https://docs.github.com/en/copilot/concepts/agents/coding-agent/about-coding-agent?ref=claudefox.com), [Third-party coding agents (GitHub Docs)](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents?ref=claudefox.com), [GitHub Copilot CLI (GitHub Docs)](https://docs.github.com/en/copilot/concepts/agents/about-copilot-cli?ref=claudefox.com), [Installing GitHub Copilot CLI (GitHub Docs)](https://docs.github.com/en/copilot/how-tos/copilot-cli/set-up-copilot-cli/install-copilot-cli?ref=claudefox.com), [Adding custom instructions for Copilot CLI (GitHub Docs)](https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-custom-instructions?ref=claudefox.com), [Adding repository custom instructions (GitHub Docs)](https://docs.github.com/en/copilot/how-tos/configure-custom-instructions/add-repository-instructions?ref=claudefox.com), [Agent skills (GitHub Docs)](https://docs.github.com/en/copilot/concepts/agents/about-agent-skills?ref=claudefox.com), [Hooks for GitHub Copilot (GitHub Docs)](https://docs.github.com/en/copilot/concepts/agents/hooks?ref=claudefox.com), [Claude plans and pricing (claude.com)](https://claude.com/pricing?ref=claudefox.com), [Claude API pricing (Claude Platform docs)](https://platform.claude.com/docs/en/about-claude/pricing?ref=claudefox.com), [How do usage and length limits work? (Claude Help Center)](https://support.claude.com/en/articles/11647753-how-do-usage-and-length-limits-work?ref=claudefox.com), [Claude Code overview (Claude Code docs)](https://code.claude.com/docs/en/overview?ref=claudefox.com), [Model configuration (Claude Code docs)](https://code.claude.com/docs/en/model-config?ref=claudefox.com), [CLAUDE.md and AGENTS.md (Claude Code docs)](https://code.claude.com/docs/en/memory?ref=claudefox.com), [Create custom subagents (Claude Code docs)](https://code.claude.com/docs/en/sub-agents?ref=claudefox.com), [Hooks reference (Claude Code docs)](https://code.claude.com/docs/en/hooks?ref=claudefox.com), [MCP (Claude Code docs)](https://code.claude.com/docs/en/mcp?ref=claudefox.com), [Plugins overview (Claude Code docs)](https://code.claude.com/docs/en/plugins?ref=claudefox.com), [Permission modes (Claude Code docs)](https://code.claude.com/docs/en/permission-modes?ref=claudefox.com), [Data usage (Claude Code docs)](https://code.claude.com/docs/en/data-usage?ref=claudefox.com), [Manage costs (Claude Code docs)](https://code.claude.com/docs/en/costs?ref=claudefox.com), [Legal and compliance: authentication and credential use (Claude Code docs)](https://code.claude.com/docs/en/legal-and-compliance?ref=claudefox.com)