xAI · tested in Taskade Genesis

How Grok builds in Taskade Genesis

Use Grok in Taskade Genesis for follow-up edits that land.

Last test

TSK-1 scorecard for Grok

TSK Score

Task completion across real apps built in Taskade Genesis

TSK Score 75 out of 100
Interfaceready to share
Strong
Taskcompletes your task
Strong
Memorykeeps your data
Strong
Adapthandles follow-up edits
Strong

Grok gets there, slowly.

More

Both of its August apps worked and both follow-up changes landed, which few models manage. The catch: it repeated failed steps dozens of times on the same requests. In October, Grok 4.7 and Grok 4.6 each built a family intake form that saved every answer and both photos, and in a later test Grok 4.7 built a reading list that saved and filtered every book.

Where does Grok rank?

TSK Score out of 100.

  1. 1GPT88 out of 100
  2. 2Claude75 out of 100
  3. 2DeepSeek75 out of 100
  4. 2Kimi75 out of 100
  5. 2Grok75 out of 100
  6. 6GLM63 out of 100
  7. 7Qwen50 out of 100
  8. 8Gemini44 out of 100
Grok versions over timeDays each version was tested
Grok versions over time: days tested so far, Aug 25 to Oct 18
DateGrok 4.7Grok 4.6Grok 4.5Grok 4.3Grok 4.20Grok Build 0.1
01, tested this day0001, tested this day
02, tested this day0001
1, tested this day3, tested this day1, tested this day1, tested this day1, tested this day2, tested this day
2, tested this day31113, tested this day

Which Grok versions did we test?

Open a version to read its findings.

Grok 4.7Tested Oct 20268 findings

The newest Grok in the Taskade model menu. In two October tests its family intake forms saved every answer and both photos and took a new question cleanly, and its reading list saved and filtered every book.

Oct 8, 2026

Built the family intake form so it saved every answer and both photos, and it refused a submission with a required answer missing.

Oct 8, 2026

The test time limit ended the build while it checked its own form, before a closing summary.

Oct 8, 2026

Added the new teacher question to the form and the teacher page, and said plainly that it had not tested a photo from a real phone.

Oct 8, 2026

Built the best Dota 2 match tracker of the Grok set: a three-page esports dashboard with labeled sample matches, a coach and a weekly automation. A logged match saved, and the heroes page landed. The time limit ended the build while it polished the look.

Oct 18, 2026

Built a reading list with stat cards, an add form and status filters. A book added from the app saved, and each filter showed the right books.

Oct 18, 2026

Built the family intake form so it saved every answer and both photos from a phone, with the parent form open and the teacher page behind a sign-in. It kept no separate column for the time of each submission.

Oct 18, 2026

Added the new teacher question, and the answer saved.

Oct 18, 2026

Added an Our Families page to an intake app that another model had built, with search, A to Z sort, a share-permission filter and a phone menu, and with contact details and teacher notes left off the cards. It kept the parent form open and named the steps the app owner must take to lock the family data.

Grok 4.6Tested Oct 20266 findings

Working apps on both August tests, and both follow-up edits landed. In September it built rich sign-up data across four pages, then added a sign-in screen nobody asked for.

Aug 25, 2026

Built a working match tracker with every asked piece, then ended without telling the user it was done.

Aug 25, 2026

Took the full 32-question sign-up form and saved every answer, but spread the data across ten projects where the leader used two.

Aug 25, 2026

Both follow-up edits landed: a ranked heroes page, and a text score field with next-step suggestions.

Aug 25, 2026

Dozens of failed build steps on each app made it one of the slowest and most expensive results of the test.

Sep 19, 2026

Took the full 32-question sign-up form into a 58-field table and four pages, with 31 of 32 questions word for word, then added a sign-in screen.

Oct 8, 2026

Built the family intake form in 14 minutes so it saved every answer and both photos, and added the new teacher question cleanly. Its teacher page showed three sample families as if they were real.

Grok 4.5Tested Oct 20263 findings

Not in the Taskade model menu. In its first test it built the most complete Grok family intake form: every answer and both photos saved, labeled sample families, a class assistant and a weekly automation.

Oct 8, 2026

Built the family intake form in 7 minutes 25 seconds so it saved every answer and both photos, with labeled sample families, a class assistant and a weekly automation.

Oct 8, 2026

Added the new teacher question to the form and to every family card, and said plainly that it had not tested the photo picker.

Oct 8, 2026

Built a four-page Dota 2 match tracker with labeled sample matches that saved a logged match and added the heroes page, but its hero chart showed no data.

Grok 4.3Tested Oct 20263 findings

Not in the Taskade model menu. Its family intake form had hard-to-read fields and no link to the teacher page, yet it saved every answer and both photos in under 4 minutes, and the new question landed.

Oct 8, 2026

Built the family intake form in 3 minutes 21 seconds so it saved every answer and both photos, but its form fields were hard to read and no link led to the teacher page.

Oct 8, 2026

Added the new teacher question to the form and the family cards.

Oct 8, 2026

Built a Dota 2 match tracker in under 5 minutes that saved a logged match and added the heroes page, but its average KDA card showed a garbled number and the board ran wider than a phone screen.

Grok 4.20Tested Oct 20263 findings

Not scored, and not in the Taskade model menu. Neither of its two October builds kept a parent's answers. A newer version can earn a fresh test.

Oct 8, 2026

With reasoning on, it described the plan, stopped before writing an app, and wrote the app only on the follow-up request. That form rejected every submission and showed no message.

Oct 8, 2026

With reasoning off, it built a dark form where the request asked for cream paper, and the saved row lost the child's name and kept only the photo file names.

Oct 8, 2026

On a Dota 2 match tracker, neither setting could log a match. With reasoning on, every save was rejected and the board showed made-up stats. With reasoning off, it built no app page, and its summary described a heroes page that did not exist.

Grok Build 0.1Tested Oct 20268 findings

Its first working apps came in October: a Dota 2 match tracker with a heroes page, a reading list, and two family intake forms. The first form dropped four of its 14 answers, and the second started the required photo permission on a choice. Every follow-up change it took in October landed.

Aug 25, 2026

Produced no projects, agents, or app: hundreds of failed build steps and no finished work.

Oct 8, 2026

Built the family intake form so the photos landed on every row, but 4 of 14 answers did not save, among them the required photo permission, while the form thanked the parent.

Oct 8, 2026

Added the new teacher question, saved it, and showed it on the teacher page.

Oct 8, 2026

Built a four-page Dota 2 match tracker with a win-rate trend that saved a logged match and added the heroes page when asked. Its sample matches carried no sample label.

Oct 18, 2026

Built the reading list, but its closing summary described the app in jargon in place of plain words.

Oct 18, 2026

Built the family intake form so it saved a parent submission with the time it was sent, but the required photo permission started on "Share with the class". The form told parents their answers were private while it linked them to a public teacher page with every phone number and email.

Oct 18, 2026

Added the new teacher question, and the answer saved.

Oct 18, 2026

Added an Our Families page behind a sign-in to an intake app that another model had built.

Model facts

The dates, limits and features of each tested model, as models.dev lists them. They describe the model, not how Taskade runs it.

Facts about each tested Grok model, from models.dev
ModelReleasedKnowledge cutoffContext window in tokensMax output in tokensInputReasoningTool callingOpen weights
Grok 4.7500K tokens500K tokensText, Images, PDFYesYesNo
Grok 4.6500K tokens500K tokensText, Images, PDFYesYesNo
Grok 4.5Not listed500K tokens500K tokensText, Images, PDFYesYesNo
Grok 4.3Not listed1M tokens30K tokensText, Images, PDFYesYesNo
Grok 4.20Not listed1M tokens30K tokensText, Images, PDFYesYesNo
Grok Build 0.1Not listed256K tokens256K tokensText, Images, PDFYesYesNo

Published specs from the model maker, via models.dev (MIT).

Published scores

The benchmark scores that labs and test sites published for each Grok model, with a link to each source. TSK-1 tests how a model builds apps. These scores test other skills.

How did Grok do on the same request?

The family intake form, three more builds

All models →
The request

Parents fill in a form on their phone with two to four photos, the teacher sees every family on one page, and then we ask for one more question on the form.

  • Grok 4.71 buildWorks

    Saved every answer and both photos from a phone, with the parent form open and the teacher page behind a sign-in. The new teacher question saved.

  • Grok Build 0.11 buildWorks, with gaps

    Saved each submission with the time it was sent, and the new teacher question saved. The required photo permission started on a choice, and parents were linked to a public teacher page with every phone number.

A reading list app

All models →
The request

A simple reading list: books with a title, an author and a status (to read, reading or done), on one page that lists them and filters them by status.

  • Grok 4.71 buildWorks

    Stat cards, an add form and status filters. A book added from the app saved, and each filter showed the right books.

  • Grok Build 0.11 buildWorks

    Built the list and its filters, but its closing summary described the app in jargon in place of plain words.

The family intake form, more models

All models →
The request

Parents fill in a form on their phone with two to four photos, the teacher sees every family on one page, and then we ask for one more question on the form.

  • Grok 4.71 buildWorks

    Saved every answer and both photos, and the new teacher question landed. The time limit ended the build before its closing summary.

  • Grok 4.61 buildWorks

    Saved every answer and both photos, and the new teacher question landed. Its teacher page showed three sample families with no sample label.

  • Grok 4.51 buildWorks

    Saved every answer and both photos, labeled its sample families, and added a class assistant and a weekly automation. The new teacher question landed.

  • Grok 4.31 buildWorks, with gaps

    Saved every answer and both photos in under 4 minutes, but its form fields were hard to read and no link led to the teacher page.

  • Grok Build 0.11 buildWorks, with gaps

    The photos landed, but 4 of 14 answers did not save, among them a required one, while the form thanked the parent.

  • Grok 4.202 builds (same result)Works, with gaps

    Neither build kept a parent's answers: one lost the child's name, and the other rejected every submission with no message.

A Dota 2 match tracker with a heroes page

All models →
The request

One player logs each match with hero, result, kills, deaths, assists, duration and notes, a dashboard shows win rate and streaks in a premium esports look, and then we ask for a heroes page.

  • Grok 4.71 buildWorks

    A three-page esports dashboard with labeled sample matches, a coach and a weekly automation. A logged match saved, and the heroes page landed.

  • Grok 4.51 buildWorks

    Four pages with labeled sample matches, a logged match saved, and the heroes page landed. Its hero chart showed no data.

  • Grok Build 0.11 buildWorks

    Four pages with a win-rate trend, a logged match saved, and the heroes page landed. Its sample matches carried no label.

  • Grok 4.31 buildWorks, with gaps

    Saved a logged match and added the heroes page, but its average KDA card showed a garbled number.

  • Grok 4.202 builds (best has gaps, worst made no app)Works, with gaps

    Neither setting could log a match: one rejected every save, and the other built no app page while its summary described one.

Most results come from one build. Where a model built the app more than once, we show its best and worst build. Read the results as a guide, not a final ranking.

How we grade these apps: read the full method.

App kits you can open today

Live apps from the official Taskade account, not builds by Grok.

Browse all App Kits →

How does Grok compare?

Other models in TSK-1

See the full TSK-1 ranking →

FAQ

What is Grok's TSK Score?

Grok scores 75 out of 100 on TSK-1: Interface Strong, Task Strong, Memory Strong, Adapt Strong. Last test Oct 18, 2026.

What is Grok best at in Taskade Genesis?

Follow-up changes. Grok 4.6 was one of the few models to land both edits we asked for on its August apps, and both apps worked. The trade-off is time: it repeated failed steps dozens of times on the same requests.

How did Grok Build 0.1 do?

In August it produced no app: it described every layer of the build and then repeated the same failed step hundreds of times until we stopped the test. In October it built a working Dota 2 match tracker, and its first family intake form lost four of 14 answers. A later October test added a reading list and a second intake form that saved each submission, though it started the required photo permission on a choice. We publish the tests that go badly alongside the ones that go well.

Can I use Grok in Taskade?

Yes. Grok 4.7, Grok 4.6 and Grok Build 0.1 are in the Taskade model menu on Max and Enterprise. Grok 4.7, the newest, saved every answer and both photos on a family intake form in its first complete test.

Where can I compare Grok with other models, or try it myself?

Side-by-side model pages live at /compare. To try the same kind of app request yourself, start from /create. This page stays on what Grok produced in TSK-1 tests.