The best ad blockers after Manifest V3, in 2026

Still the most capable blocker in existence — on Firefox. Several of its strongest filter types need APIs only Firefox provides, and Mozilla has committed in writing to keeping them, with at least 12 months’ notice if that ever changes. 10,444,513 average daily Firefox users, 4.80 stars, v1.72.2 from 8 July 2026. The catch is Chrome-specific and dated: it is a Manifest V2 extension, Chrome has had MV2 disabled since 2025, and the listing leaves the Chrome Web Store on 31 August 2026. It was still live, with a working install button, when we checked on 29 July 2026.

The same author’s Manifest V3 blocker, free with no in-app purchases, and the right answer if you are staying on Chrome. In the PoPETs 2026 study it matched full uBlock Origin on ordinary display ads. Since June 2026 it supports custom filters and remote filter-list subscriptions, which needs the per-extension “Allow User Scripts” switch turned on — not on by default, so flip it. Two honest limits: generic cosmetic filtering is off by default here (it is on in full uBO), and its author says plainly that it is too different from uBO to be an automatic replacement.

The best all-round pick across browsers: 16,000,000 Chrome users at 4.7 stars and 1,682,636 average daily Firefox users, and the first to work around MV3’s limits — custom filters returned in v5.2.77 on 22 September 2025 and filter lists began updating independently of the extension in v5.4 on 7 May 2026. It also keeps a separate, still-live MV2 listing with about 30,000 users for holdouts, and it is the only vendor here that spans extension, Safari app and DNS.

Nothing about Safari was damaged by Manifest V3 — Safari’s content blockers have been declarative since 2015 and get 150,000 rules each, roughly five times Chrome’s guaranteed budget. Wipr 2 is the cleanest way to buy into that: $4.99 once, 4.68 stars from 1,240 ratings, updated 29 July 2026. Its in-app purchases are tips plus an optional network add-on, not a paywall. It makes no YouTube claim, which we read as restraint rather than weakness.

The Safari blocker to buy if you want lifetime access and the most control: free to download, then $2.99 a month, $14.99 a year, or $39.99 once for lifetime. It carries the largest rating base of any Safari blocker we checked — 4.50 from 12,725 ratings — and was updated on 25 May 2026. It is one of the few whose developer claims YouTube blocking, but that is the developer’s own claim, scoped to Safari, and no developer claims to block ads inside the native YouTube app.

The open-source alternative, and a long-standing MV3 critic that ships an MV3 build anyway: 2,000,000 Chrome users at 4.7 stars, 972,094 average daily Firefox users. Date its criticism correctly — the substantive post is from December 2021, before its own MV3 release. One correction, because it matters: its blog says uBlock Origin is already gone from Chrome’s store, and that is false — the listing was live when we checked on 29 July 2026. Its Safari app is free but has only 194 ratings, too thin to recommend on that platform.

The largest install base in the category — 37,000,000 Chrome users — but read this before installing: its Acceptable Ads programme is ON BY DEFAULT, and eyeo charges a licensing fee to large platforms that benefit commercially from it. Turning it off takes two clicks: the toolbar icon, the gear, General, then uncheck Acceptable Ads. An independent committee governs the criteria. Technically it is also the least elegant MV3 build here: uBlock Origin’s author criticises it for force-restarting its service worker roughly every 30 seconds when idle.

Not a general ad blocker — the EFF’s tracker-specific tool, which learns what is following you rather than reading a filter list, and it retains observational webRequest under MV3. Run it alongside a blocker, not instead of one. A telling data point: it has more average daily users on Firefox (1,761,720) than it has users on Chrome (1,000,000), a rare inversion that suggests the privacy-conscious already moved. Its MV3 criticism, like Ghostery’s, dates from December 2021 — the EFF’s March 2025 post does not mention MV3 at all.