AI Compare
Find the Right AI for What You Do
Choose your tasks, set your priorities, and find AI tools that fit your work and budget.
Prices for United States, before sales tax
- 1Pick what you use AI forOne or more. You can change it later.
- 2Tell us what mattersPriorities, must-haves and your budget.
- 3Get your best matchesUp to three plans, with prices and sources.
What do you want to do with AI?
Select every card that applies to you, then press Next.
- Research & Search Sources that hold up · Fact checks · Research reports
- Writing & Content Emails · Article edits · Long drafts
- Images & Design Images from text · Photo edits · Text in images · Logos
- Coding & Development Code questions · Editor agents · Pull requests · App builders
- Video & Animation Clips from text · Footage edits · Presenters · Captions
- Business & Productivity Google Workspace · Connected apps
More uses
- Data & Analytics Spreadsheet answers · Charts from data
- Agents & Automation Browser tasks
- Audio & Voice Songs · Voiceover · Transcripts · Dictation
LLMs & APIs. API and model selection needs workload-based costs and its own evaluations. It is not part of this tool yet.
Find my AI runs in your browser and needs JavaScript. Nothing you choose leaves your device. The plan list and the comparisons below work without it.
Compare two AIs
Pick two plans. Every row says what each plan documents, when we last checked it, and where it comes from.
Price
$20 a month
$19.99 a month or $199.99 a year
Models and limits
GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 Sol/Terra/Luna, legacy models; GPT-6 Astra limited in Work/Codex.
3.8 Flash; 3.1 Pro (Expanded); no Deep Think.
GPT Instant 54K (~40 pages); Reasoning 256K (~320 pages).
1 million tokens.
Pricing page model list for every individual plan shows only OpenAI models (GPT-6.1 Sol, GPT-6 Astra, GPT-6 Sol, GPT-6 Luna, GPT-5.6 family, legacy). No models from other AI companies are listed.
Only inside Antigravity: Pro column lists Gemini models plus Claude Sonnet 5.5/Opus 5.5 (non-trial subscriptions only), Claude 4.6 models and GPT-OSS-120b (4.6/GPT-OSS removed Nov 2, 2026). Gemini app lists Google models only.
ChatGPT chat assistant for everyday text chats (writing, questions, everyday tasks); pricing table lists everyday text chats as Unlimited* on Free, Go, Plus and Pro, subject to abuse guardrails. Plus adds advanced reasoning models, projects, scheduled tasks, custom GPTs.
Gemini app with 4x Free usage limits, 3.8 Flash and 3.1 Pro, Deep Research, plus Gemini in Gmail, Docs and Sheets.
Research and sources
Search the web; pricing table marks Search on all plans.
Search grounding with source links not described on fetched plan pages.
Expanded deep research.
Expanded Deep Research.
Files and data
Expanded messages and uploads.
Up to 1,500 pages of file uploads.
Data analysis Yes; interactive tables and charts.
Spreadsheet/CSV computation not described on fetched plan pages.
Coding
Codex included; limited GPT-6 Astra usage, optional credits.
Jules (higher limits), Antigravity (entry limits), AI Studio, Android Studio.
Codex IDE extension (signs in with ChatGPT). Editors: VS Code, VS Code Insiders, Cursor, Windsurf (Codex extension); Xcode and JetBrains IDEs via their own integrations. Usage shared with Codex/ChatGPT Work limits.
Via Google Antigravity: Antigravity IDE and CLI, plus extensions for VS Code, Visual Studio, JetBrains, Zed, Xcode. Pro: higher quota refreshed every 5 hours up to a weekly limit, plus AI-credit overages. Code Assist IDE/Gemini CLI consumer access ended June 18, 2026.
Codex IDE extension is documented as an agent beside your code (chat with editor context, review diffs, delegate to cloud). No official page fetched documents inline as-you-type completions, and none says they are absent.
Antigravity baseline (all plans) includes unlimited Tab completions; Pro adds higher agent rate limits, not more completions.
Codex Cloud runs coding tasks on repos in OpenAI-managed cloud environments (continues while your computer sleeps); you can commit or open a pull request when ready. Eligible Plus, subject to rollout; usage counts toward plan limits.
Jules in Pro: 100 daily tasks (rolling 24h), 15 concurrent tasks, higher access to latest model (Jules limits page); cloud VM tasks that can open PRs. Gemini subscriptions page: 'Higher limits'.
Codex code review on GitHub PRs: comment @codex review or turn on automatic reviews; posts a GitHub review flagging P0/P1 issues; optional Security Review (research preview). Pricing matrix marks GitHub PR reviews available for Plus. Uses plan usage.
Consumer Gemini Code Assist on GitHub (which covered Google AI Pro users) shut down July 17, 2026; Code Assist IDE/CLI also stopped serving Google AI Pro on June 18, 2026. No replacement PR reviewer documented for Pro.
Images, video and voice
Image generation incl. 'with Thinking'.
Expanded image generation (Nano Banana Pro).
No video generation listed on pricing page; Sora web/app discontinued.
Omni Flash; 1,000 Flow credits/mo; limited Veo 3.1 Lite trial.
Voice Expanded; voice with video.
Gemini Live.
Image editing and design
ChatGPT Images (all tiers) edits images you upload or it made: describe the change, or select an area in the editor to add, remove or update parts. Plus also gets Images with thinking.
Gemini app (Nano Banana 2): upload an image and ask Gemini to make edits, change parts of it, or combine several photos; editing is 18+. Plan adds Nano Banana Pro ('Redo with Pro') and higher limits.
ChatGPT Images can follow an instruction to make an image's background transparent, and you can upload an existing image to edit. No separate background remover tool.
Google says you can tell Gemini to 'change the background, replace an object, or add an element'; no page documents removing a background to transparency.
OpenAI says ChatGPT Images can follow instructions to add text; its poster template tip says to describe any text you want included.
Nano Banana 2 lists 'Better text rendering: Add text to images with more accurate spelling'; Nano Banana Pro (paid plans) is pitched for images with text rendering.
Images has templates by category, named as a starting point for a poster or logo (Images > Templates). Templates are not yet in Work mode.
No logo maker, logo templates or logo mode documented; Google's Nano Banana page says you can create logos, invites and posters with clear text. No dedicated logo tool or template set is documented.
No OpenAI page documents SVG or other vector output for ChatGPT Images; downloads are described only as saving the image.
No Google page documents SVG or vector output for Gemini images; downloads are full-size images (2K with a Google AI plan, 1K without).
No OpenAI page documents upscaling in ChatGPT Images. The mobile image controls include Resize, which is not described as upscaling.
The Google AI plans comparison shows no 4K image upscaling (Google Flow) for this plan; it is ticked only for Ultra 5x and Ultra 20x.
Video editing and presenters
No video tool on any ChatGPT plan: the pricing comparison has no video row, and Sora (OpenAI's video app, which animated images) is closed.
Gemini app (Gemini Omni): turn up to 5 photos into a video. Also Google Flow Frames to Video (1,000 Flow credits/month plus 50 daily) and Google Photos Photo to video (Expanded generations).
No video editing on any ChatGPT plan: the pricing comparison lists no video feature and the Sora app, OpenAI's only video tool, is closed.
Gemini app and Google Flow (Gemini Omni): prompt edits to uploaded or generated video (Flow: upload up to 60 s, 1 GB, edit a 10 s segment, 40 credits per edit; Scenebuilder trim and arrange). Google Vids: text-based transcript trim. Uploading video for edits is not offered in some US states.
No OpenAI page documents a digital presenter feature for ChatGPT; OpenAI has no video product since Sora closed.
Google Vids AI avatars speak the words in your script (25/month, includes personal avatars). Gemini app and Flow also let you record your own face and voice as an avatar for Omni videos (not in EEA, UK, Switzerland; English only in Gemini).
No OpenAI page documents lip sync for ChatGPT; OpenAI has no video product since Sora closed.
No Google page documents lip sync to new audio for these plans. Google says of Gemini Omni: editing video to change audio and speech is still being tested.
No OpenAI page documents automatic video captions in ChatGPT.
Not stated as automatic: Google Vids has styled captions that sync with the audio, open to personal accounts; no page says they are generated without your input. Google Vids styled captions, synced to the audio, available to all personal Google accounts (also transcript trim and teleprompter); you can correct caption errors.
No OpenAI page documents turning long videos into short clips in ChatGPT.
No Google page ties automatic long video to short clips to the Gemini app, Flow, Vids or Photos on this plan.
ChatGPT makes no video on this plan (no video row on the pricing page, Sora closed), so a video watermark does not apply.
Visible watermark can be turned off: Gemini app (Settings, Media, Watermark) and Google Flow have a visible watermark toggle in the US. Invisible SynthID watermark and C2PA Content Credentials always remain.
App building and hosting
ChatGPT Sites (public beta): describe a website, app or game in Work (web) or Work or Codex (desktop app) and ChatGPT builds it, with a private preview. Building or editing counts toward Codex usage.
Canvas in the Gemini app creates apps and web pages from a prompt, with code view and Gemini-powered features. Google AI Studio Build (vibe coding agent, 'Expanded' daily usage on this plan) builds full-stack web and Android apps from a prompt.
Sites publishes to a ChatGPT-hosted URL; you choose selected people, workspace or public. No fixed limit on the number of Sites on Plus or Pro, but usage and storage limits apply; hosting usage is separate from Codex usage.
Canvas apps can be shared with a public g.co/gemini/share link (opens in gemini.google.com). AI Studio deploys apps to Cloud Run with a public URL; needs a Cloud billing account except the Starter Tier (eligible accounts, up to 2 apps).
Sites lets you rename the ChatGPT-hosted URL or connect an apex domain or subdomain you own via DNS records (Site settings > Add domain). Sites does not register domains. Help says 'Where custom domains are available'.
No Google page documents putting a Canvas or AI Studio app on your own domain. AI Studio lets you pick a custom subdomain under ai.studio (for example your-app-name.ai.studio).
Sites offers D1 relational database (10 GB per site) and R2 object storage for files, plus 'Sign in with ChatGPT' so visitors can save their work.
Canvas apps can save data across sessions and devices (anyone with a public link can view and edit that data). AI Studio Build provisions Firebase Firestore and Firebase Authentication (Sign in with Google).
Sites can publish a compatible local project from the ChatGPT desktop app or Codex; for a local source project each saved version is tied to its Git commit. No download of a web-built Site's code is documented.
Canvas has a Code view where you can read and edit an app's code, and exports Python to Colab. Google AI Studio Build exports code as a ZIP or syncs two ways with GitHub.
Music, voice and transcription
No official ChatGPT page found that describes making songs with vocals; help center search for music returned no such feature.
Lyria 3.5 in the Gemini app makes tracks with music and lyrics (vocals) from text, photos or video; short (about 1 min) or full (2 to 3 min, needs the Pro model); 18+. Ticked on every plan; higher plans get higher limits. Google Vids 'Generate music' is also included.
No official ChatGPT page found that describes instrumental music generation.
When creating a track you can choose whether it has vocals or not, so instrumental tracks are possible; same usage limits as music generation.
No official ChatGPT page describes stem downloads; no music generation feature documented.
Help article lists downloads as MP4 video with cover art or audio-only MP3; separated stems are not described.
ChatGPT Voice speaks ChatGPT's own replies in chosen voices, but no official page describes reading text you supply as a downloadable voiceover.
Google Vids 'AI voiceover' and Google Docs 'Audio generation' are both listed for this plan.
No official ChatGPT page describes copying a voice; Voice offers a fixed set of preset voices (Arbor, Breeze, Cove, and others).
Not documented: Google's personal avatar feature records your face and voice for videos, but no page says it reads new text in a copy of your voice.
No official page describes translating and re-voicing audio or video files. Voice can be asked to speak another language in a live chat only.
Not documented for recorded audio or video: Google Meet translates speech live in calls on some plans, which is not dubbing a file.
No official ChatGPT page describes generating sound effects.
No official page describes generating stand-alone sound effects from a description in the Gemini app or Google Flow.
Audio uploads (WAV, MP3, M4A, FLAC and more, up to 512 MB) can be transcribed; transcripts may contain errors. Record on the macOS app also live-transcribes sessions up to 4 hours.
Audio files can be uploaded to Gemini for answers and summaries (total audio up to 10 minutes; 3 hours with Google AI Pro or Ultra), but no official page names transcription as a feature.
ChatGPT Record (macOS desktop app) records meetings or voice notes, transcribes, and writes summary notes as a canvas; up to 4 hours per session. Calendar auto-join works with Google Meet links only.
Gemini 'Take notes for me' in Google Meet captures notes in Google Docs, gives 'Summary so far' and a recap email; recommended for 15 minutes to 8 hours; 8 spoken languages.
Dictation (microphone icon) turns speech into editable text inside ChatGPT's own message box only; no official page describes typing into other apps.
No official Gemini page describes system-wide dictation into other apps; help center search found only voice input inside Gemini.
Projects and memory
Projects, shared projects, scheduled tasks.
Gemini Notebook (research/writing assistant, notebooks).
Custom GPTs (use); new GPT creation not available on personal accounts per Free FAQ.
Gems ('Custom AI experts').
Expanded memory and context.
Memory of past chats; requires 18+, personal account, Keep Activity on. Plan not specified.
Integrations and agents
Plugins; apps connecting to internal tools; developer mode (beta).
Connected Apps: Gmail, Calendar, Maps, YouTube, YouTube Music, Google Home, Flights, Hotels, Spotify, WhatsApp; varies by device/country.
No ChatGPT add-on that works inside Gmail is documented; Gmail data reaches ChatGPT only through its own connectors and plugins.
All Gemini in Gmail features in Google's compare table, incl. Help me write, Proofread, AI Overviews (threads and Gmail search, search US only), AI Inbox, Help me schedule. US Pro/Ultra users may no longer see the Gmail side panel; in-line AI replaces it.
No ChatGPT add-on that works inside Google Docs is documented; OpenAI lists add-ins for Word, PowerPoint, Excel and Google Sheets only.
Gemini in Docs: Help me write, Write and edit, Summarize a document, Create personalized documents, Generate an image, Audio generation. Docs Live (voice) rolling out on Android and iOS in US only. Select languages, 18+.
ChatGPT for Google Sheets add-on: build, analyze and update sheets inside Sheets. Pricing table: Yes for Plus.
Gemini in Sheets: AI function, Enhanced Smart Fill, Generate insights and analyze data. Works best with native Sheets files; Excel (.xlsx) files must be saved as Google Sheets first.
ChatGPT for Word add-in: draft, edit and format inside Word. Pricing table: Yes for Plus.
No official Google page found that documents a Gemini add-in for Microsoft Word on any personal Google AI plan. Only third-party add-ins turned up. Gemini app can export files, which is not working inside Word.
ChatGPT for Excel add-in: build, analyze and update sheets inside Excel. Pricing table: Yes for Plus.
No official Google page found that documents a Gemini add-in for Microsoft Excel on any personal Google AI plan. Only third-party add-ins turned up. Gemini app can export files, which is not working inside Excel.
ChatGPT for PowerPoint add-in: create and edit slides inside PowerPoint. Pricing table: Yes for Plus.
No official Google page found that documents a Gemini add-in for Microsoft PowerPoint on any personal Google AI plan. Only third-party add-ins turned up. Gemini app can export files, which is not working inside PowerPoint.
No ChatGPT add-in that works inside Outlook is documented; the Outlook Email plugin works from ChatGPT's own chat, which is a connector.
No official Google page found that documents a Gemini add-in for Microsoft Outlook on any personal Google AI plan. Only third-party add-ins turned up. Gemini app can export files, which is not working inside Outlook.
ChatGPT Work agent on desktop, web and mobile; built-in browser.
Chrome auto browse (US and India) and Gemini Spark (select countries).
Privacy and control
Default ON; turn off 'Improve the model for everyone' in Settings > Data controls.
Default ON for 18+; turn off Keep Activity (chats still kept up to 72 hours).
Request a copy of data from settings or Privacy Portal.
Export via Google Takeout.
No ads.
No official statement found about ads in the Gemini app.
Rights and licensing
OpenAI Terms of Use (individual services, all plans): you own the Output and OpenAI assigns its rights in it to you; Usage Policies still apply and Output may not be unique.
Not stated as a permission: Google's Terms say Google won't claim ownership of generated content and set no plan-specific ban; no page grants commercial use in so many words.
Apps and extras
No cloud-storage bundle.
5 TB.
Web, iOS, Android; desktop app on macOS and Windows.
Web, Android, iOS, desktop app for Windows 10+ and macOS 15+ (Apple Silicon).
Plan a project
Each stage of the project, the app and plan that documents it, and how the work moves from one stage to the next.
Project plans run in your browser and need JavaScript. Nothing you choose leaves your device.
Check my subscriptions
Add what you pay for. See the monthly and yearly cost, the next charges, and what you would lose if you dropped one.
This tool runs in your browser and needs JavaScript. Nothing you enter leaves your device.
Head-to-head comparisons
Written by our desk from the same data, with the trade-offs and the tests to run. Pick a pair.
| ChatGPT | Claude | Gemini | Perplexity | X Subscription (Cursor, Grok, X) | GitHub Copilot | |
|---|---|---|---|---|---|---|
| ChatGPT | vs | vs | vs | vs | vs | |
| Claude | vs | vs | vs | vs | vs | |
| Gemini | vs | vs | vs | vs | vs | |
| Perplexity | vs | vs | vs | Not compared: a research product and a coding tool do different jobs | Not compared: a research product and a coding tool do different jobs | |
| X Subscription (Cursor, Grok, X) | vs | vs | vs | Not compared: a research product and a coding tool do different jobs | vs | |
| GitHub Copilot | vs | vs | vs | Not compared: a research product and a coding tool do different jobs | vs |
20 other products have no written comparison yet. Compare any two side by side, from the same documented facts.
- ChatGPT vs Gemini: which plan fits your work?
- Claude vs Gemini: which plan fits your work?
- ChatGPT vs Claude: which plan fits your work?
- Claude vs Perplexity: which plan fits your work?
- ChatGPT vs Perplexity: which plan fits your work?
- Gemini vs Perplexity: which plan fits your work?
- Cursor vs GitHub Copilot: which plan fits your work?
- ChatGPT vs Cursor: which plan fits your work?
- ChatGPT vs GitHub Copilot: which plan fits your work?
- Claude vs Cursor: which plan fits your work?
- Claude vs GitHub Copilot: which plan fits your work?
- Gemini vs Cursor: which plan fits your work?
- Gemini vs GitHub Copilot: which plan fits your work?
Every plan we compare
89 individual plans from 26 products, priced for United States, before sales tax. 5,518 facts, each with its official source.
ChatGPTOpenAI
- ChatGPT Free Free
- ChatGPT Go $8 a month
- ChatGPT Plus $20 a month
- ChatGPT Pro 100 $100 a month
- ChatGPT Pro 200 $200 a month
- ChatGPT Pro 500 $500 a month
ClaudeAnthropic
- Claude Free Free
- Claude Pro $20 a monthor $200 a year
- Claude Max 5x $100 a month
- Claude Max 20x $200 a month
GeminiGoogle
- Gemini (free) Free
- Google AI Plus $4.99 a month
- Google AI Pro $19.99 a monthor $199.99 a year
- Google AI Ultra 5x $99.99 a month
- Google AI Ultra 20x $199.99 a month
PerplexityPerplexity AI
- Perplexity Free Free
- Perplexity Pro $20 a monthor $204 a year
- Perplexity Max Price not verifiedor $2,004 a year
X Subscription (Cursor, Grok, X)Anysphere
- X Subscription Free Free
- X Subscription Plus $8 a monthor $80 a year
- X Subscription Premium $30 a monthor $300 a year
- X Subscription Super $100 a monthor $1,000 a year
- X Subscription Ultra $200 a monthor $2,000 a year
GitHub CopilotGitHub (Microsoft)
- Copilot Free Free
- Copilot Pro $10 a month
- Copilot Pro+ $39 a month
- Copilot Max $100 a month
SunoSuno, Inc.
- Suno Free Free
- Suno Pro $10 a monthor $96 a year
- Suno Premier $30 a monthor $288 a year
UdioUncharted Labs, Inc.
- Udio Free Free
- Udio Standard $10 a monthor $96 a year
- Udio Pro $30 a monthor $288 a year
ElevenLabsElevenLabs
- ElevenLabs Free Free
- ElevenLabs Starter $6 a monthor $60 a year
- ElevenLabs Creator $22 a monthor $220 a year
- ElevenLabs Pro $99 a monthor $990 a year
SuperwhisperSuperUltra, Inc.
- Superwhisper Free Free
- Superwhisper Pro $8.49 a monthor $84.99 a year
RunwayRunway AI, Inc.
- Runway Free Free
- Runway Standard $15 a monthor $144 a year
- Runway Pro $35 a monthor $336 a year
- Runway Max $95 a monthor $912 a year
LumaLuma AI, Inc.
- Luma Plus $30 a monthor $300 a year
- Luma Pro $90 a monthor $900 a year
- Luma Ultra $300 a monthor $3,000 a year
HiggsfieldHiggsfield, Inc.
- Higgsfield Free Free
- Higgsfield Starter $19 a monthor $228 a year
- Higgsfield Plus $59 a monthor $564 a year
- Higgsfield Ultra $129 a monthor $1,188 a year
HeyGenHeyGen Technology, Inc.
- HeyGen Free Free
- HeyGen Creator $29 a monthor $288 a year
- HeyGen Pro $49 a monthor $488 a year
SynthesiaSynthesia Limited
- Synthesia Basic Free
- Synthesia Starter $29 a monthor $216 a year
- Synthesia Pro $89 a monthor $768 a year
DescriptDescript, Inc.
- Descript Free Free
- Descript Hobbyist $24 a monthor $192 a year
- Descript Creator $35 a monthor $288 a year
MidjourneyMidjourney, Inc.
- Midjourney Basic $10 a monthor $96 a year
- Midjourney Standard $30 a monthor $288 a year
- Midjourney Pro $60 a monthor $576 a year
- Midjourney Mega $120 a monthor $1,152 a year
IdeogramIdeogram, Inc.
- Ideogram Free Free
- Ideogram Plus $20 a monthor $180 a year
- Ideogram Pro $60 a monthor $504 a year
Adobe FireflyAdobe Inc.
- Firefly Free Free
- Firefly Standard $9.99 a month
- Firefly Pro $19.99 a month
- Firefly Pro Plus $49.99 a month
- Firefly Premium $199.99 a month
CanvaCanva US, Inc. (Canva Pty Ltd group)
- Canva Free Free
- Canva Pro $18 a monthor $144 a year
LovableLovable Labs Incorporated
- Lovable Free Free
- Lovable Pro $25 a monthor $250 a year
ReplitReplit, Inc.
- Replit Starter Free
- Replit Core $20 a monthor $216 a year
- Replit Pro $100 a monthor $1,080 a year
Microsoft CopilotMicrosoft
- Microsoft Copilot (free) Free
- Microsoft 365 Personal $9.99 a monthor $99.99 a year
- Microsoft 365 Family $12.99 a monthor $129.99 a year
- Microsoft 365 Premium $19.99 a monthor $199.99 a year
- Microsoft 365 Pro $99.99 a month