Article
CommunityJev Media & Talks
AICrier: Celeris-1 Decision hits 80% on JevBench
AICrier's news-feed summary of Celeris-1 Decision: a multimodal diffusion model for bounded structured decisions that Celeris says reaches 80% on JevBench, 7 points above Jev 1.13.0, with 67 ms median latency.
- Format
- Article
- Reading time
- 2 min
- Published by
- Community
- Author
- tom_w_hamer
- Last verified
- Oct 2026
Highlights
- The feed item reports Celeris-1 Decision at 80% on JevBench, 7 points above Jev 1.13.0, with 67 ms median latency.
- AICrier calls the target gap compelling: fast typed decisions application code can consume directly, between classifiers and general-purpose LLMs.
- System One API compatibility is flagged as letting existing Jev integrations switch with minimal code changes.
- The notes list multimodal inputs expanding use cases to receipt checks, document review, visual verification and agent control.
- AICrier cautions developers to validate calibration, edge cases and benchmark methodology on their own workloads.
Watch out
Copy reviewed against the published AICrier feed item (8 October 2026) on 2026-10-17. AICrier is an aggregator: the item summarizes Celeris's own announcement, so the 80% JevBench figure and 67 ms latency are Celeris's claims, and AICrier's bullet analysis is editorial commentary rather than independent testing.
Questions about this entry
Short answers drawn from the linked project or page.
It is described as a multimodal diffusion model for bounded, structured decisions over text, JSON and images, returning typed outputs that application code can consume directly.
80% accuracy on JevBench, which the item says is 7 points above Jev 1.13.0, with 67 ms median latency, all attributed to Celeris rather than to independent measurement.
It says Celeris targets a compelling gap between classifiers and general-purpose LLMs: fast, typed decisions that integrate directly into application code without parsing and routing overhead.
The item says existing Jev integrations can switch to Celeris with minimal code changes because the API convention is compatible, so clients already written for decision endpoints can be repointed.
AICrier lists receipt checks, document review, visual verification and agent control, alongside real-time routing, voice systems and high-volume automation where the 67 ms latency claim is most relevant.
It calls the JevBench lead promising but says developers should validate calibration, edge cases and benchmark methodology on their own workloads rather than trusting the headline figure.
The item is credited to tom_w_hamer and was published 2026-10-08, tagged as a model-release announcement sourced from Celeris's product channels and rated 9/10 for relevance on the AICrier feed.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: Simon Willison's overview of Noul, Choice, and Score, and where a decision model fits.
benwerd53 pointsJev introduces a new shape of LLMRead the thread on Hacker NewsHacker News: swyx posts the Latent Space interview with Diogo Almeida.
swyx4 pointsJev: System One Models for Prod, Not God – With Diogo Almeida, CEO, TypeSafe AIRead the thread on Hacker NewsX: Diogo Almeida in five minutes
X: LLMs vs. Jev, clearly explained
More Jev resources like this
Fortune reports Jev went viral among developers, with nearly 13% of Vercel's paid teams trying it within a day of its gateway arrival. Three weeks after Jev's launch, OpenAI rolled out a competing Decisions API built on GPT-6 Luna.
Jev Media & TalksArticle#community#press#competition
Community
LangChain's September 25 post argues Jev's typed decisions fit LangGraph's deterministic orchestration, with a litigation review example and a Stagehand browser case where act() latency fell to 0.46 seconds.
Jev Media & TalksArticle#community#press#explainer
Community
A playground operator's audit of every published Jev number: TypeSafe's four-workflow evals, the favourable homepage demo, independent tests and its own calls. It concludes the price and speed are real and the intelligence claim is roughly mid-tier frontier on decision workflows.
Jev Media & TalksArticle#community#press#explainer
Community
Adrian Bridgwater's September 21 report on TypeSafe's launch pitch: a model that does not write but decides, a new architecture and RLCD training, and Vercel's Jev adoption numbers through its AI Gateway.
Jev Media & TalksArticle#community#press#explainer
Community
Simon Willison released llm-typesafe 0.1a0, an LLM plugin for TypeSafe's Jev: install with llm install llm-typesafe, set a typesafe API key, then ask noul yes/no questions, Choice questions and Score questions from the command line. A demo refund question returned noul 0.99.
Jev Media & TalksArticle#community#press#explainer
Community
TNW reports Jev reached nearly 13% of Vercel's paid teams within 24 hours, the AI Gateway's fastest adoption: twice the GPT-5.6 family's share and more than 6x Fable 5.1's, at $0.042 per million input tokens with output unmetered.
Jev Media & TalksArticle#community#press#adoption
Community
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.