ChatGPT

Claude vs ChatGPT vs Gemini vs Copilot

Compare vendor-documented prices, limits, context windows, data controls, and accuracy guidance for Claude, ChatGPT, Gemini, and Copilot without claiming that any assistant produced a better result.

Sources checked 3 Oct 2026

For a Claude vs ChatGPT vs Gemini comparison with Copilot as the fourth option, vendor pages settle current plan prices, documented features, usage rules, context windows, and stated data controls. They do not establish which assistant will handle your work better, and Chat Picker has not tested these assistants for quality, speed, accuracy, reliability, or benchmark performance.

What the vendors document: Claude vs ChatGPT vs Gemini

ChatGPT. As read on October 1, 2026, OpenAI's ChatGPT pricing page lists Plus at $20 per month. Its Free Tier FAQ, read the same day, says Free includes unlimited everyday text chats with GPT-5.6 Luna, while separate limits apply to uploads, image generation, voice, data analysis, and GPT use. It also says ads may appear in certain countries. The pricing page lists app context windows of 27K for Free, 54K for Go and Plus, and 128K for Pro for instant models; reasoning models are listed at 256K for Go and Plus and 400K for Pro. The page lists a training opt-out for Free, Go, Plus, and Pro, while the same-day ChatGPT Business overview says Business workspace data is not used for training. OpenAI's accuracy note, also read that day, says ChatGPT can be incorrect or misleading and advises checking important information.

Claude. Anthropic's plans and pricing page, as read on October 3, 2026, lists Free at $0, Pro at $17 per month with annual billing or $20 billed monthly, and Max from $100. Its app context row, checked on October 1, lists up to 1M on every plan, depending on the model. The October 3 page says usage resets on a rolling five-hour window, paid plans also have weekly limits, and additional weekly, monthly, model, or feature limits may apply. Free includes web search, file creation, and code execution; Pro adds more usage and tools. Model training is listed as opt-out. Anthropic's incorrect-response guidance, read on October 1, says Claude can produce convincing but incorrect statements and should not be your only source of truth.

Gemini. Google's AI plans page, as read on October 1, 2026, lists Free at $0 with a Google Account, Google AI Plus at $4.99 per month, AI Pro at $19.99, and Ultra at $99.99 or $199.99. Google's limits page, read the same day, lists context windows of 32K without an AI plan, 128K with Plus, and 1M with Pro or Ultra. It says limits refresh every five hours until the weekly limit is reached. The plans page lists 2x Free usage for Plus and 4x for Pro, plus storage of 15 GB, 400 GB, and 5 TB respectively. Google's source-display guidance says a Sources button appears when sources are available, but the Gemini material read does not state a training control or a direct warning that answers may be wrong.

Microsoft Copilot. Microsoft's Microsoft 365 plans for individuals, as read on October 1, 2026, lists Personal at $9.99 per month, Family at $12.99, Premium at $19.99, and Pro at $99.99. The first three include a one-month trial and automatically renew unless canceled; Pro is monthly and also renews automatically. Personal includes 1 TB of storage and Copilot in Word, Excel, PowerPoint, and Outlook desktop apps. Microsoft's business pricing page, read the same day, lists chat connected to work data, Microsoft Teams, more than 10 additional apps, and Enterprise Data Protection with management and compliance controls. Its Copilot help page says to check, improve, and build on generated content. That page links to data-use and protection information but does not state a specific Copilot chat-retention period or model-training choice.

What the documentation cannot tell you

Plan descriptions do not show whether an assistant will correctly interpret your files, follow your instructions, calculate spreadsheet values, surface useful sources, request the right permissions, or give consistent results across repeated tasks. Those questions depend on your documents, prompt, selected feature, account settings, and review process.

Chat Picker's method page says comparison figures carry a vendor source and the date it was read. When a figure cannot be confirmed, the site marks it “Not verified” and leaves the cell blank. This rewrite removes older statistics and test results that had no sources.

How to check it yourself

Test research and source use. Give each assistant: “Explain why daylight saving time changes the clock. Cite primary sources and flag any statement that depends on a country.” Open the cited pages, check whether they support the claims, and record unsupported statements, missing qualifications, and broken links.
Test document grounding. Upload the same policy document and ask: “State the renewal and cancellation rules in this document. Quote the exact lines and write ‘Not stated’ for anything the document does not answer.” Record invented terms, omitted exceptions, and quotations that do not match the source.
Test spreadsheet work. Give each assistant the same redacted expense report and ask: “Recalculate every total, show the formulas, and flag mismatches without changing the source data.” Check the formulas and source cells, then record each correction needed.
Test your workflow. In each assistant's native workspace, upload the same meeting notes and ask: “Create an action list with owner, due date, and status; mark missing fields instead of guessing.” Record permission requests, formatting changes, export steps, and manual edits.
Check limits in context. Use a representative session rather than a deliberate stress test. Record the plan, billing basis, date and time, region, model or feature label, prompt and file sizes, limit notices, reset information, and any charge shown before you subscribe.

Which rows of the comparison matter

Open the Claude vs ChatGPT vs Gemini comparison matrix and prioritize these rows:

Plan type, price, and billing basis
Free-plan access
Usage caps and reset rules
Context window, uploads, and storage
Web search and source display
Training and data controls
Business seats, administration, and security
Integration with the applications you already use

Do not treat bundled Microsoft 365 prices as directly equivalent to individual assistant subscriptions. For Copilot, use the Microsoft plan pages to check what your bundle includes and how its data terms apply to your organization.