proomt

Search

Search posts, papers, and topics

All posts

Hacker News front page8 min readrelease notesintermediate

Laya the open source version of Jev

Read on convaiinnovations.comHN 1326·313 commentsLobsters 2

Summary

Laya is an open‑source, bidirectional‑encoder model family for ultra‑fast, calibrated decision‑making (choice, score, boolean) over structured schemas. It runs 6‑8× faster than the closed‑source Jev, supports 100+ languages via three checkpoints, and includes a lightweight router that selects the appropriate checkpoint before inference. Benchmarks show higher accuracy, far better calibration (ECE…

  • System‑1 style models (non‑autoregressive) can replace generative LLMs for reflex decisions, delivering calibrated probabilities without hallucination.
  • Three decision primitives (choice, score, noul) simplify API design and guarantee well‑formed outputs.
  • Routing by script detection enables a single hub repository to serve English, multilingual, and typed‑decision checkpoints with negligible overhead.
  • Empirical results: Laya outperforms TypeSafe Jev on accuracy, calibration, latency, and cost; however, choice tasks with many options (>20) still suffer.

Many production pipelines waste compute and money calling large generative LLMs for binary or multi‑class decisions that require only a calibrated score. Laya demonstrates a practical, open alternative that can be run on commodity GPUs, dramatically reducing latency and cost while providing trustwo…

8/10
llmreinforcement learningperformanceml infraopen sourcemultilingual

Related reading

  2. Introducing System One Models and Jev

    TypeSafe AI announced its first “System One” model, Jev, a non‑text‑generating LLM that outputs type‑safe structured decisions with calibrated probabilities. It claims 40‑200× lower latency (70‑500 ms) and 100‑500× lower cost versus frontier LLMs, no hallucinations, and parallel sampling. The post includes a side‑by‑side demo, a custom “workflow” benchmark comparing Jev to GPT‑5.6/6 and other mod…

    Hacker News front pagetypesafe.ai9 minHN1824480lobste.rs26

  3. Reverse-engineered Jev-like model

    Jevlike is an open‑source starter model that scores a list of text options in a single forward pass. It provides a minimal architecture (option queries, shared dot‑product scorer), synthetic data generation, training/evaluation CLI, and examples on Doom and chess. The repo supports a byte‑level encoder or a frozen Hugging‑Face encoder (e.g., Qwen2.5‑0.5B), runs on CPU/MPS/CUDA, and reports benchm…

    Hacker News front pagegithub.com4 minreleaseHN16224

  4. TypeSafe AI's Jev now available on AI Gateway

    Vercel AI Gateway now offers Jev, a probabilistic decision model that returns typed choices, scores, and booleans instead of raw text. TypeSafe AI reports it runs up to 193× faster and 445× cheaper than standard LLMs, exposed via the experimental evaluate API in AI SDK 7.

    Vercelvercel.com2 minrelease

  5. OpenJev

    OpenJev is a browser‑only demo that lets you load small LLM checkpoints (e.g., MiniCPM‑5 2B, Qwen3 0.6B) onto your GPU and compare two inference paths: reading raw logits for a set of options versus prompting the model to emit a JSON with option probabilities token‑by‑token. The page reports model sizes, download times, balanced accuracy on a few benchmarks, and wall‑clock timings measured with `…

    Hacker News front pageopenjev.com2 minreleaseHN709288

  6. Jev is the fastest-adopted model in AI Gateway history

    Vercel reports that the Jev decision model was adopted by 13% of paid teams within its first day, outpacing prior model launches. Jev claims to be up to 194× faster and 445× cheaper than general‑purpose LLMs while returning structured, probabilistic decisions.

    Vercelvercel.com1 minHN21