Claude vs Grok

As of Nov 6, 2026, Claude is clearly ahead of Grok on measured coding and reasoning, while Grok wins on real-time X posts, image and video generation, and price. Claude Opus 5.5 scores 59 on the Artificial Analysis Intelligence Index against 35 for Grok 4.7, and Claude Code with Opus 5.5 scores 64.85% on Terminal-Bench 4.0 against 37.58% for Grok Build with Grok 4.7. Grok now comes in The X Subscription, from $8 a month with X Premium and Cursor bundled, against $20 for Claude Pro. xAI, Grok's maker, has been part of SpaceX since Feb 2, 2026 and now signs itself SpaceX Super Intelligence (SpaceXSI) (xAI joins SpaceX).

Sources: Claude plans and pricing, xAI pricing, Introducing The X Subscription, Introducing Grok 4.7, Grok, all read Nov 6, 2026.

Learn more: Grok on ai.vi.ki and ChatGPT vs Grok on chatgpt.vi.ki

Verdict by task

Coding

Claude wins by a wide margin on the public boards. On Terminal-Bench 4.0, Claude Code with Claude Opus 5.5 scores 64.85% and Grok Build with Grok 4.7 (xhigh) 37.58%; on WebDev Arena, Opus 5.5 is first at 1810 and Grok 4.7 21st at 1637. xAI's own launch table agrees on coding: it shows Claude Fable 5.1 ahead of Grok 4.7 on CursorBench 4.0 (51.8% against 46.3%) and Terminal-Bench 4.0 (57.9% against 37.6%) (Introducing Grok 4.7). Grok's plans do include Cursor, which itself can run Claude at API prices (xAI pricing); see Claude Code vs Cursor.

Reasoning and knowledge

Claude wins. The Artificial Analysis Intelligence Index gives Opus 5.5 (max) 59 and Grok 4.7 (xhigh) 35; ARC-AGI-2 gives Opus 5.5 93.3% and Grok 4.7 61.4%; Artificial Analysis's Humanity's Last Exam runs give 61.4% against 43.1%. On Arena's text board, Opus 5.5 is third at 1509 and xAI's best entry, Grok 4.20 Beta 1, is 40th at 1475. xAI's table has one row where Grok 4.7 leads: the Harvey Legal Agent Benchmark, 19.6% against 6.7% for Fable 5.1 (Introducing Grok 4.7). See Claude Benchmarks.

Writing

Close. In Tom's Guide's Mar 25, 2026 test of the same five prompts, Claude won three rounds and Grok two, with Grok praised as more genuine in two; its headline said "the winner wasn't obvious" (Tom's Guide). That test predates both current models, so tone is a matter of taste.

News and real-time information

Grok wins. It searches the web and X live, with "Real-time X integration" for breaking news and trends, and it runs inside X itself (Grok). Claude's Web Search reads the open web with citations on every plan but has no special access to X posts (Web search). Zapier also names Grok the chatbot "for X integration" (Zapier).

Images and video

Grok wins outright. Grok Imagine makes images (Imagine Image 2.0, eighth on Arena's text-to-image board) and video, with 1080p, 15-second clips on the Super plan (xAI pricing). Claude generates neither (Can Claude produce images?; see Image Generation).

Voice

Grok has the newer offering: Grok Voice 3.0 launched on Oct 14, 2026 (xAI news). Claude's Voice Mode is a beta on every plan with a limited set of voices (Use voice mode). No independent test compared the two.

Price

Grok wins. The X Subscription starts at $8 a month (Plus), and Premium at $30 bundles the benefits of SuperGrok, Cursor Pro and X Premium+, which xAI says cost $90 combined before (Introducing The X Subscription). Claude Pro is $20 a month, or $17 billed annually, and Claude Max and Claude Ultra match Grok's Super and Ultra at $100 and $200. One caveat: Grok's usage is a single weekly pool shared with Cursor (xAI pricing). On the API, Grok 4.7 costs $2 / $6 per million tokens against $4 / $20 for Opus 5.5 (Claude pricing).

Office and work tools

Even. Grok has add-ins for Word, PowerPoint, Excel and Outlook and one for Google Workspace (Jun to Jul 2026), plus Connectors, Skills and Automations (xAI news). Claude has Claude for Microsoft 365, Claude for Google Workspace, Connectors and Skills.

Privacy

Not scored here. xAI's privacy pages refused automated reading on Nov 6, 2026, so Grok's training setting could not be confirmed from its own source. Claude trains on consumer chats unless you opt out in Settings > Privacy, and never uses Incognito Chats for training (Claude Privacy and Data Use).

When to choose Grok

Choose Grok if you follow news and trends on X, want images and short videos, or want the most in one bundle: X Premium, Cursor and an assistant for $30. Developers who need a capable model at a low token price may also prefer Grok 4.7 at $2 / $6 per million tokens.

When to choose Claude

Choose Claude for coding, analysis and long documents, where the measured gap is large: 64.85% against 37.58% on Terminal-Bench 4.0 and 59 against 35 on the Artificial Analysis index. Anthropic and SpaceXSI are not only rivals: on May 6, 2026 Anthropic agreed to use xAI's Colossus 1 data center to add capacity for Claude Pro and Max subscribers (SpaceXSI). For other options, see Claude Alternatives.

What reviewers found

2:02 In an Aug 26, 2026 video, Peter Yang compared ChatGPT, Claude, Grok and Gemini on ten everyday uses, with Claude Opus 5 and Fable 5 on Anthropic's side, models since replaced; he did not name Grok's model. Asked to prototype a recipe app, Claude Code built more working features, but he found its look "very Claude generated" and missing images, while Grok's prototype surprised him as the most visually impressive, with generated images that matched the recipes.

7:17 He still gave design overall to Claude, counting Claude Design, which asked him clarifying questions before drawing, and the slide decks and videos Claude made for him, and called Grok "a pretty good contender" for websites and apps.

24:35 On personal agents, he found Grok Bot cleaner to use than ChatGPT's app and liked that each Grok agent gets its own cloud computer, but found it too restrictive for now; he said Claude Code was "playing catch-up" with ChatGPT on browser use, voice and mobile.

Frequently asked questions

Is Claude better than Grok? For coding and reasoning, yes: as of Nov 6, 2026, Claude Opus 5.5 scores 59 on the Artificial Analysis index against 35 for Grok 4.7. Grok is better for live X posts, images and video, and price.

Is Grok cheaper than Claude? Yes. The X Subscription starts at $8 a month and Premium at $30 includes X Premium and Cursor, while Claude Pro costs $20. Grok 4.7's API price is also lower than Claude Opus 5.5's.

Can Claude see X posts like Grok? No. Claude searches the open web with citations, while Grok searches X live and runs inside X.

Can Claude make images and videos like Grok? No. Claude does not generate images or video; Grok Imagine does both.

Which is better for coding, Claude or Grok? Claude. Claude Code with Opus 5.5 scores 64.85% on Terminal-Bench 4.0 against 37.58% for Grok Build with Grok 4.7, and xAI's own table shows Claude Fable 5.1 ahead on CursorBench and Terminal-Bench.

Sources

• ChatGPT vs Claude vs Grok vs Gemini: The Best AI for 10 Use Cases (August 2026), Peter Yang, Aug 26, 2026 (video)
• Plans and pricing, Anthropic, read Nov 6, 2026
• Can Claude produce images?, Claude Help Center, read Nov 6, 2026
• Use voice mode, Claude Help Center, read Nov 6, 2026
• Web search, Claude Help Center, read Nov 6, 2026
• xAI joins SpaceX, SpaceXSI, Feb 2, 2026
• New Compute Partnership with Anthropic, SpaceXSI, May 6, 2026
• Introducing Grok 4.7, SpaceXSI, Sep 21, 2026
• Introducing The X Subscription, SpaceXSI, Oct 12, 2026
• Pricing, SpaceXSI, read Nov 6, 2026
• Grok, SpaceXSI, read Nov 6, 2026
• News, SpaceXSI, read Nov 6, 2026
• Terminal-Bench 4.0 leaderboard, tbench.ai, read Nov 6, 2026
• WebDev Arena, Arena, read Nov 6, 2026
• Text Arena, Arena, read Nov 6, 2026
• Text-to-Image Arena, Arena, read Nov 6, 2026
• LLM Leaderboard, Artificial Analysis, read Nov 6, 2026
• Humanity's Last Exam Benchmark, Artificial Analysis, read Nov 6, 2026
• ARC Prize Leaderboard, ARC Prize Foundation, read Nov 6, 2026
• I tested Grok vs. Claude with the same 5 prompts, Tom's Guide, Mar 25, 2026
• The best AI chatbots, Zapier, read Nov 6, 2026