How we compare tools
The comparison method: checked figures, job groups, and no scores before a hands-on test.
Which figures get printed
Each tool row comes from a vendor check dated 2026-10-11. A paid amount is printed only when that check was complete and the price sentence itself was not marked as still to be confirmed. A free-tier line is printed on the same rule. A paid trial is not relabeled as a free plan.
Segments that were still unconfirmed are dropped, even if a neighboring segment on the same vendor had a readable price. That is why a suite price can appear while a separate toolkit price stays blank. Partial checks stay blank even when a number was visible in the source note.
The date on a row is the check date, not the day the site was built. If a vendor changes a plan after that date, the page can be wrong until the next check. The cell does not pretend otherwise.
How tools are grouped
Tools are grouped by the job a reader is trying to finish: text-to-video, avatar, clips, editing, voice, SEO briefs, general drafting, grammar, fiction, charting, scanning, journals, research, or no-code stock automation. The group is the job the tool is used for.
Inside a group, the default order is editorial. You can sort by name or by a checked monthly figure. Tools without a checked figure sort to the end of a price sort. Filters hide rows. They do not delete them from the page, so the full set is still in the HTML.
Why there is no score
A score would require a published hands-on protocol and a session that follows it. This version has the protocol outline on the how-we-test page and does not have the sessions. Stars stay hollow, and the structured data has no rating.
Headings that say "best" match the way people search. The body says the order is not a finished ranking. A one-line fit on a head-to-head page names a common use.