Five frontier model drops in nine days and nobody has a real benchmark for any of them. when the…
five frontier model drops in nine days and nobody has a real benchmark for any of them. when the cadence moves faster than the evals, the eval chart stops being a chart and starts being a launch announcement.
that should worry you.
Context
First-party pages exist for several of the launches around 1 to 3 September 2026: Anthropic's Claude Fable and Mythos 5.1 page, Google's Gemini 3.8 Flash blog from 2 September, Meta's Muse Spark 1.3 post from 2 September and OpenAI's GPT-6 Astra release on 3 September.
Later secondary write-ups count the launches differently. One says five major models in the ten days from 1 to 10 September, another says five in three days, and another says five in ten days. All were published after the note.
The sources disagree on the window, so five in nine days is not confirmed. That nobody has a real benchmark for any of them is the author's claim. Later comparisons exist but I did not read them.
Watch next
- First-party model cards for launch-day evaluation numbers.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 4 September 2026 at 11:06 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →