Grubby AI vs Walter Writes AI

Both ladders are priced in words a month, so the arithmetic runs cleanly for every tier that states a pool. Price is reported on this site and never scored.

Grubby AI figures read on its pricing page on 9 November 2026; Walter Writes AI figures read on its homepage pricing block the same day. The like-for-like row is 30,000 words: Grubby Pro at $119.88 a year against Walter Starter at $96, with the same three modes and the same 1,500-word cap on Grubby's side against 750 on Walter's. Two caveats attach. Walter's monthly toggle prices were not in the page as read, so the month-to-month column runs one way. And Grubby's page served two different plan sets to two requests on 9 November 2026, one of them listing 500 free words and three unnamed tiers at $47.88, $107.88 and $168.00 a year; this page reports the version that names its plans and caps, and a buyer should confirm the price on the page before paying.

Grubby's Essential plan caps an input at 500 words, Pro at 1,500 and Unlimited at 2,500, with a 30,000-word daily fair-use line on Unlimited, read 9 November 2026. Walter's Starter caps a request at 750, Pro at 1,500 and Elite at 2,000, read 9 November 2026. At the entry tier Walter's cap is 250 words larger; at the middle tier they match; at the top Grubby's is 500 words larger.

Splitting matters because each piece is rewritten without sight of the others, and the joins are the writer's job afterwards. For the student Grubby's page addresses, a 1,500-word essay is three pastes on the advertised plan and one on the $9.99 plan. The published method uses inputs of 350 to 500 words, so every plan on both ladders accepts them whole; the cap is scored under limits and access, 15 points of 100.

Both vendors name Simple, Standard and Enhanced, read on both pages on 9 November 2026. Walter describes its three as a range from subtle to advanced and lists them on every plan. Grubby gives Essential buyers Simple only and adds Standard and Enhanced from Pro, so the mode ladder is a reason to buy up rather than a control every buyer has. Neither page defines what each mode changes, and a shared name is no evidence that the rewrites are alike.

The method runs each tool on its default setting and on its strongest documented mode as a second row, so a published run will show both tools twice. For Grubby that means Simple and Enhanced on a plan that includes both; for Walter it means the mode its interface opens on and Enhanced.

No peer-reviewed study has tested either product by name in a way this site could cite, so the research here is about the category. Weber-Wulff and colleagues tested 14 detectors and found every one below 80 percent accuracy, with accuracy falling further on paraphrased and machine-translated text (Weber-Wulff et al. 2023, International Journal for Educational Integrity, read 9 November 2026).

Grubby's pricing page leads with a money-back guarantee tied to Turnitin's AI detector and repeats claims of a 100 percent human score on every plan tile, read 9 November 2026, with no input set, detector version or date. Turnitin's detector is not available to the public, so no vendor can show a Turnitin reading a reader can reproduce. What Turnitin publishes is that its AI writing checker helps identify when AI bypassers, which it also calls humanizers, may have been used in a submission (Turnitin, AI writing detection page, read 9 November 2026). A refund is a commercial term, not a measurement. Walter Writes AI ships a built-in detector that returns what it calls an authenticity score after each rewrite, read 9 November 2026; a vendor's detector reading its own output is not independent evidence either.

Those sources measure detector readings, not whether the rewritten text kept the writer's claims. The method on this site weights meaning kept at 35 points of 100, errors introduced at 20, detector pass rate at 30 and limits and access at 15, and two editors rate every output blind. When a run is published, this page gains meaning, error and per-detector fields for both tools. No detector score should be the only evidence in a misconduct case.