What did The Rundown report about Jev's launch?

TypeSafe, the startup from former OpenAI researcher and ChatGPT contributor Diogo Almeida, announced early access to Jev on September 15. The model answers preset questions inside software, per the brief.

What three kinds of questions can Jev handle?

According to TypeSafe's workflow evaluations cited by the piece: yes/no questions with probabilities, choosing among defined options, and scoring an input against a scale, with a confidence measure returned for the choice and scoring formats.

What does the piece say about the can't-hallucinate claim?

TypeSafe says Jev can't hallucinate because it only selects from allowed answers, but The Rundown notes that guarantee concerns the answer's format: a model can select an allowed option and still make the wrong judgment.

What price comparisons does the article give?

TypeSafe advertises $0.042 per million input tokens, or $42 per billion, with free output. The piece compares that with Claude Fable 5.1's listed base input price of $10 per million tokens, about 238 times Jev's rate.

What speed does TypeSafe claim for Jev?

The report cites responses in 70 to 500 milliseconds and speedups of 40 to 200 times for structured decision queries, with latency tests generally run near its West Coast service, and notes the company acknowledges added time and cost in its probability method.

How does the piece suggest developers use confidence scores?

Per TypeSafe's recommendation: automatic action for clear cases, confirmation or more information when confidence drops, and escalation for uncertain cases, with developers testing accuracy and setting conservative thresholds on their own data.

What open questions does the piece flag?

Response times under load and sustained pricing remain open questions, and the piece says a developer weighing Jev must measure the whole workflow, including people who review uncertain cases and the consequences of a wrong decision.

Article

CommunityJev Media & Talks

The Rundown: TypeSafe launches Jev for AI decisions in software

The Rundown's news brief reports TypeSafe's September 15 early access launch of Jev, a model that answers preset questions inside software, and flags that the company's can't-hallucinate guarantee concerns format, not correctness.

See highlights

Open therundown.ai

Format: Article
Reading time: 2 min
Published by: Community
Author: The Rundown Editorial Team
Last verified: Oct 2026

Tagscommunity press calibration adoption

Works forJev for Classification Jev for Decision gating

Used inJev for Software & DevTools

Highlights

TypeSafe announced early access to Jev on September 15; The Rundown pitches it for sorting requests, scoring records and jailbreak screening.
Per TypeSafe's workflow evaluations, Jev answers yes/no questions with probabilities, chooses among defined options, or scores an input against a scale.
TypeSafe says Jev can't hallucinate because it only selects from allowed answers; the piece notes that guarantee is about format, not correctness.
TypeSafe advertises $0.042 per million input tokens with free output, about 238 times below Claude Fable 5.1's listed $10 per million base input price.
TypeSafe reports responses in 70 to 500 milliseconds and speedups of 40 to 200 times, with latency tests generally run near its West Coast service.

Watch out

Copy reviewed against the published article on 2026-10-17. Price, speed and reliability claims are TypeSafe's; the piece notes the evaluation reference answers average judgments from GPT-6 Astra and Claude Fable 5.1, so agreement gives only a partial view of correctness.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

The weekend builds: 838 projects and the claims behind them

Oct 16, 2026

The madewithjev gallery counted 838 builds over Jev's first 28 days. A walk through its sampled X posts: what gets built first, the costs authors report, and why every number needs its source.

Jev Media & Talks#community#use-cases#adoption

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai