TSK-Bench · Build arena
Which AI model builds the best intake form?
Every model got the same request, word for word, in Taskade Genesis. Here is how each one did, best first.
In our tests, Claude Opus 5, Claude Opus 5.5, DeepSeek V4 Pro, DeepSeek V4 Pro 0813, DeepSeek V4 Pro, max thinking, DeepSeek V4.1 Flash, GLM 5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, and Grok 4.7 all built a working app. Each result comes from one to five builds, so read it as a direction, not a final rank.
Last test · 38 builds
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.
What does “works” mean here?
- The app opens and runs from the one request, with no fix from us.
- We use it the way a customer would: we fill in its form or log an entry, then check that every answer was saved.
- It covers what the request asked for. "Works, with gaps" means it runs but misses part of the request.
Where the request includes a follow-up change, the result covers the build and the change together.
How did each model rank?
Works first, then works with gaps, then no app. Each result is one group, with its models listed by name.
| Model | |||
|---|---|---|---|
| Works · 13 | |||
Claude Opus 5 Saved every answer and both photos, and its email automation ran. The follow-up change landed. Works1 build · Tested | Works | 1 | |
Claude Opus 5.5 Saved every answer and both photos, turned away an oversized photo, and opened in the cream look the request asked for. The follow-up change landed. Works1 build · Tested | Works | 1 | |
DeepSeek V4 Pro Saved every answer and both photos, and the new teacher question landed. Two sample families showed as real. Works1 build · Tested | Works | 1 | |
DeepSeek V4 Pro 0813 Both settings saved every answer and both photos, and the new teacher question landed. With max thinking, the follow-up message finished the build. Works2 builds (same result) · Tested | Works | 2same result | |
DeepSeek V4 Pro, max thinking The time limit ended the build. The follow-up message finished a form that saved every answer and both photos, with no made-up families. Works1 build · Tested | Works | 1 | |
DeepSeek V4.1 Flash Saved every answer and both photos, and its new-family automation ran. Three sample families showed as real. Works1 build · Tested | Works | 1 | |
GLM 5 Saved every answer and both photos, and the new teacher question showed on its gallery page, but no link led to that page. Works1 build · Tested | Works | 1 | |
GPT-5.6 Luna All five saved every answer and both photos, and the new teacher question saved and showed on the teacher page each time. Only one saved the date and time of each submission that the request asked for, and one teacher page showed raw answer codes. Works5 builds (best works, worst has gaps) · Tested | Works | 5best works, worst has gaps | |
GPT-5.6 Sol Saved every answer and both photos, labeled its sample family, and kept the teacher page open with no sign-in. The new teacher question landed. Works1 build · Tested | Works | 1 | |
GPT-6.1 Sol All three saved every answer and both photos, and every follow-up change landed. One teacher page misread its saved answers. Works3 builds (same result) · Tested | Works | 3same result | |
Grok 4.5 Saved every answer and both photos, labeled its sample families, and added a class assistant and a weekly automation. The new teacher question landed. Works1 build · Tested | Works | 1 | |
Grok 4.6 Saved every answer and both photos, and the new teacher question landed. Its teacher page showed three sample families with no sample label. Works1 build · Tested | Works | 1 | |
Grok 4.7 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. Works1 build · Tested | Works | 1 | |
| Works, with gaps · 11 | |||
Claude Haiku 4.5 Saved every answer, but the family photo replaced the child photo on every submission. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
Claude Sonnet 4.5 Saved every answer and both photos in a plain look, and its open teacher page listed parent phone numbers. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
Claude Sonnet 5 All three saved every answer and both photos, and the new teacher question saved each time. None saved the date and time of each submission that the request asked for, and one teacher page showed raw answer codes. Works, with gaps3 builds (same result) · Tested | Works, with gaps | 3same result | |
DeepSeek V3.2 Neither setting saved a family: one save was rejected, and the other sent the form to an address that does not exist. Works, with gaps2 builds (same result) · Tested | Works, with gaps | 2same result | |
DeepSeek V4 Pro, high thinking The form dropped every photo a parent picked, so the two-photo rule blocked every submission. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
GLM 4.7 Saved every answer and both photos, but the teacher page crashed on load. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
GPT-5.6 Terra All three builds saved every answer and both photos in about 4 minutes or less. One opened in a dark look where the request asked for cream paper. Works, with gaps3 builds (best works, worst has gaps) · Tested | Works, with gaps | 3best works, worst has gaps | |
Grok 4.3 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. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
Grok 4.20 Neither build kept a parent's answers: one lost the child's name, and the other rejected every submission with no message. Works, with gaps2 builds (same result) · Tested | Works, with gaps | 2same result | |
Grok Build 0.1 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. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
Mistral Large 4 A polished form, but it dropped every photo a parent picked, so the two-photo rule blocked every submission. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
| No app · 1 | |||
| No app | 1 | ||
Earlier tests (8)
The same request, word for word, in an earlier test of the same versions.
| Model | |||
|---|---|---|---|
Claude Sonnet 5 Saved every answer and both photos, and its teacher page showed each answer choice in plain words. The new teacher question saved. Works1 build · Tested | Works | 1 | |
Grok 4.7 Saved every answer and both photos, and the new teacher question landed. The time limit ended the build before its closing summary. Works1 build · Tested | Works | 1 | |
Claude Sonnet 5 Both builds saved every answer and both photos, and the new teacher question landed. Sample families showed on the teacher page with no label. Works2 builds (same result) · Tested | Works | 2same result | |
GPT-5.6 Luna One build saved every answer and both photos in under 5 minutes. In the other, every parent submission failed on the photo permission question. Works, with gaps2 builds (best works, worst has gaps) · Tested | Works, with gaps | 2best works, worst has gaps | |
Grok Build 0.1 The photos landed, but 4 of 14 answers did not save, among them a required one, while the form thanked the parent. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
Claude Sonnet 5 Saved every answer and both photos. The follow-up change landed. Works1 build · Tested | Works | 1 | |
GPT-5.6 Luna A complete build: every answer and both photos saved. Its public assistant told a visitor a child's name. Works, with gaps1 build · Tested | Works, with gaps | 1 | |
DeepSeek V4.1 Flash Saved every answer and both photos on a two-page form, and the follow-up change landed. The build ran past the time limit. Works, with gaps1 build · Tested | Works, with gaps | 1 |
Each row is one model version: its newest test on this request, the number of builds behind it, and what we found. Rows with the same result form one group and are listed by model name. A direction, not a final rank.
Follow-up changes to this app
More Build arena tests
No side-by-side results published yet: Sales CRM, Inspection app, AI governance tracker, Habit tracker app, and Recipe app.
The Build arena →The full TSK-Bench ranking →Pick a model in TaskadePlans and pricingDownload the data (JSON)
How we grade these apps: read the full method.
FAQ
Which AI model builds the best intake form?
In our tests, Claude Opus 5, Claude Opus 5.5, DeepSeek V4 Pro, DeepSeek V4 Pro 0813, DeepSeek V4 Pro, max thinking, DeepSeek V4.1 Flash, GLM 5, GPT-5.6 Luna, GPT-5.6 Sol, GPT-6.1 Sol, Grok 4.5, Grok 4.6, and Grok 4.7 all built a working app. Each result comes from one to five builds, so read it as a direction, not a final rank.
How many AI models were tested on this intake form request?
Models tested: 23, from 7 model families. Each got the same request, word for word. Some ran at more than one setting, for 25 results in all. 13 worked, 11 worked with gaps, and 1 did not produce an app.
Can I build an intake form in Taskade?
Yes. Describe the intake form in Taskade Genesis in your own words, pick a model or let TSK-1 Auto choose, and ask for changes after the first build.
