CommunityJev Media & Talks
A decision model d1 surpasses Jev
GIGAZINE's report on Liquid AI's d1, which it describes as the first decision model to beat Jev on Hugging Face's Decision Index, and a short survey of the clone wave that followed Jev's September launch.
- Format
- Article
- Published by
- Community
- Last verified
- Oct 2026
Works forJev for Evaluation
Used inJev for Software & DevTools
Highlights
- Reports Liquid AI's claim that d1 is the first model to beat Jev on the Decision Index.
- Explains Jev as a decision-only model from Diogo Almeida with near-zero output cost.
- Places d1 within the wider wave of models built on Jev's ideas.
Watch out
The benchmark comparison is Liquid's own run of the Decision Index; the article is a news report of that claim, not an independent evaluation.
Questions about this entry
Short answers drawn from the linked project or page.
That Liquid AI released d1, a decision model with functionality similar to Jev, and that Liquid says d1 was the first model to outperform Jev on Hugging Face's Decision Index.
Liquid AI's own run of the Decision Index 0.2.1 tests, in which d1 scored higher than Jev on many of the rows. The article presents the comparison as published by Liquid, not an independent rerun.
As a model from Diogo Almeida, a major ChatGPT contributor, that outputs a judgment with a confidence score instead of text, with high speed and near-zero output cost.
That d1 was available through the Liquid AI API and was free at the time of writing, with pricing to be announced later. Terms change quickly, so check the current console.
Because d1 is presented as one of many models built on Jev's ideas. The article frames the burst of similar decision models as the direct consequence of Jev's launch.
It is a vendor run on the vendor's own execution of a public benchmark. Read the per-axis numbers and your own task results rather than the headline order.
GIGAZINE, a Japanese technology news site, which also covered Jev's own launch. The English edition reaches an international audience.
Related terms
Glossary definitions related to this entry.
More Jev resources like this
A three-week timeline of the Jev clone war: the September 15 stealth launch, reported $10B valuation talks, an Amazon engineer's own build, and the $870M round, alongside OpenAI, Amazon, and Cloudflare shipping rivals.
Jev Media & TalksArticle#community#news#funding
Community
Turing Post TV tests Jev from early access in the TypeSafe playground, starting from the constraint that it does not write essays or code.
Jev Media & TalksVideo#community#video#evaluation
Community
India Jade explains RLCD as a response to using a frontier model for a small decision, which is the job TypeSafe trains Jev for.
Jev Media & TalksVideo#community#video#explainer
Community
ML Visualized retells Diogo Almeida's Latent Space point that a launch chart with the new model on top of every row is not something you can trust as a benchmark.
Jev Media & TalksVideo#community#video#evaluation
Community
IBM's 25 September 2026 Mixture of Experts episode spends a segment on Jev, setting TypeSafe's speed and cost claims against how the company built its benchmarks.
Jev Media & TalksArticle#community#podcast#ibm
Community
Prism Labs checks the launch demos against TypeSafe's own footnotes: what they keep from the 70 ms claim, and what they drop from the speed multiples.
Jev Media & TalksVideo#community#video#evaluation
Community
From the Jev community
Posts from builders shipping with Jev right now.