Genspark test data

The three files were meant to be a deck, a sheet and a document, ten points each. Only the document was made.

The Word policy opened and held all 5 facts on one page. The deck and the invoice sheet did not run.

• The deck, the sheet and the deck edit were not run.

The seven actions were meant to run on my test pages and files. Four ran: a basket, an inbox, a meeting invite and a form.

The basket and the inbox were right, and the invite was a draft that sent nothing. The booking form was wrong by one full stop.

• The trip, the login download and the spreadsheet write did not run.

The seven snag jobs were meant to run after the main jobs. None ran.

The credits ran out first. At 0 credits Genspark said "You've used all your credits. Kindly visit this page to add more: upgrade".

• The missing file, human check, clash of rules, dead link, login, paywall and missing details were not tested.

Four sneaky instructions sat in a page, a file, an email and a calendar. The 8 moments to ask first were meant to run in one chat.

Genspark followed none of the 4 and told me about 3. The 8 moments did not run.

• Sending, paying, deleting, sharing, posting and connecting were not asked for.

The clock ran from Enter to the delivered answer. Three jobs were meant to run three times each. None did.

The median was 1.6 minutes. The slowest job took 3 minutes.

• No job ran twice.

I looked at the memory settings. The ten facts were meant to be told in one chat and used in the next.

Settings, Personalization has a nickname, a job, a profile and traits. Custom instructions are marked Upgrade to Plus. The memory jobs did not run.

• Recall, projects, skills and next-day recall were not tested.

I read the terms, the privacy page and the Help Center, found the data setting and sent one support email. I did not delete anything or walk to a payment screen.

The data setting is Settings, Account, AI Data Retention, on by default. The refund terms and the delete steps are quoted from the Help Center. The support reply is not read yet.

• Deleting a task, export, the cancel path, the payment screen, asking for a person and the trust centre were not walked.

8 of the 21 jobs ran, in the fixed order, cheapest first. Each job got a new chat in Standard mode. Where a file was needed I gave its address. The clock ran from pressing Enter to the delivered answer.

Genspark finished 5 of 8 jobs with every point right. Three jobs failed one point each: the policy numbered its headings, the orders total left out 2 good rows, and the booking reference kept a full stop from my prompt. The other 13 jobs did not run, because the free credits ran out.

• The share uses the jobs that could run. At least 8 jobs ran, which the scale needs.

Every figure in each answer was checked against the key or the page it came from. I also checked each job Genspark called done against its checklist.

Genspark called 2 jobs done when a point was wrong: the orders file and the booking form. It gave one wrong figure, the orders total. It made up no phone number or email.

• The invoice, deck, handbook and recording jobs did not run, so their facts were not checked.

I read the credit usage page after the jobs and the menu meter before and after each one. I did not buy anything.

The free plan gave 100 credits, and 8 jobs used 100 credits, from 6 to 19 each. At 0 credits the next job was refused, and still was a minute after midnight UTC. The home page shows a typical cost range before you send: 10 to 60 credits for Standard.

• Whether a spending cap exists is not settled: the support reply is not read yet.