The most interesting model this week doesn't write text at all. jev, out of stealth sept 15 from…
the most interesting model this week doesn't write text at all. jev, out of stealth sept 15 from ex-openai rlhf researcher diogo almeida, prices input at $0.042 per million tokens and returns calibrated probabilities over answers you define in one parallel pass, with no decoder and nothing to hallucinate.
the llm-first stack looks like a detour for narrow decisions.
Context
TypeSafe AI's post, dated 14 September 2026, introduces Jev, available in early access, with input priced at 0.042 dollars per million tokens and output listed as free. It describes a parallel sampler trained with Reinforcement Learning for Calibrated Decisions that returns type-safe structured values with calibrated probabilities over answers you define, and gives up string generation. The post says it runs in 70 to 500 milliseconds and is 40x to 200x faster, and the author, Diogo Almeida, writes that he helped build instruction-following methods at OpenAI.
The page is dated 14 September and the note says 15 September, a one-day gap that was not resolved. The speed and cost figures come from TypeSafe's own workflow evals made by its own team, which the page says may carry some bias, so they are vendor-reported and not independent. The phrase no decoder was not found in the text read. Nothing to hallucinate is a vendor claim for a model that emits only predefined answer types, which limits free-text errors and does not show the answers are right or the probabilities calibrated beyond the vendor's own evals. Almeida's OpenAI background is his own statement. Jev is a specialised model sold as a function call, not a general language model. The line that an LLM-first stack looks like a detour for narrow decisions is the author's opinion.
Watch next
- Independent calibration tests and pricing after early access.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 20 September 2026 at 01:19 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 →