CommunityJev Models & Reimplementations
Liquid AI d1
Liquid AI's first decision model, a hosted API that adds image input to the Choice, Score, and Noul shape. Liquid reports it matching or beating GPT-6.1 Sol on four of six tested applications at 19x to 200x lower cost.
- Published by
- Community
- Last verified
- Oct 2026
Highlights
- Answers yes/no, choice, and score questions with probabilities and supports text or image input.
- Liquid reports matching or beating GPT-6.1 Sol on four of six tested applications at 19x to 200x lower cost.
- Open-weight d1-3B and d1-omni-600M are on Hugging Face under the LFM Open License v1.0.
- Text decisions are reported at 200 to 300 ms in one forward pass, with no output tokens.
Watch out
The comparative and Decision Index numbers come from Liquid AI itself; confirm quality and calibration on your own task before switching, and check the open-model licenses separately.
Questions about this entry
Short answers drawn from the linked project or page.
Liquid AI's first decision model, served from its API. It answers yes/no, choice, and score questions with probabilities and adds image input, so it can make structured decisions about pictures as well as text.
Liquid reports d1 matching or beating GPT-6.1 Sol on four of six tested applications at 19x to 200x lower cost, and says d1 was the first model to outperform Jev on Hugging Face's Decision Index.
Yes. d1 supports text or images, or both, as input and makes decisions in one forward pass without generating tokens. Text decisions are reported at 200 to 300 ms.
Liquid released open-weight d1-3B and d1-omni-600M on Hugging Face under the LFM Open License v1.0, with day-one llama.cpp support, separate from the hosted API and its vision model.
Through the Liquid console and API. Because it returns typed decisions rather than chat text, it uses a decision endpoint in the same spirit as Jev's /v1/systemone rather than /chat/completions.
No. They come from Liquid AI and from its own run of the Decision Index. Treat the claims as vendor results and check the axes that matter to you, since a headline score can hide a calibration or latency trade-off.
When image input is required, when you want an open weight you can self-host, or when the hosted d1 benchmarks better on your task. Jev stays the text-first reference with the widest tooling and gateway support.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
More Jev resources like this
A multimodal decision model on Workers AI that takes text, images, audio, or video in one request and returns schema-constrained probabilities, built on a 30B mixture-of-experts backbone with about 3B parameters active.
Jev Models & Reimplementations#community#cloudflare#multimodal
Community
17GitHub stars
Jev-compatible decision models with vision input: LoRA fine-tunes of Qwen3.5 at 4B and 9B serving Choice, Score and Noul on the /v1/systemone wire format, with screenshots and photos placed beside text in the state. Runs on your own NVIDIA GPU; weights are non-commercial.
Jev Models & Reimplementations#community#vision#open-weights
Community
Xiaooolong/vevPython
Xiaooolong/vevPython110GitHub stars
Ascend NPU fork of the Laya decision checkpoints that answers Jev-style Choice, Score and Noul questions in a single forward pass on Huawei 910B hardware. It reports 37-47 ms median for a four-question request, 33.8x-70.9x faster than one container CPU thread.
Jev Models & Reimplementations#community#ascend-npu#decision-model
Community
zzhdbw/laya-AscendPython
zzhdbw/laya-AscendPython9GitHub stars
A Jev-style typed-decision baseline built on the frozen Qwen3.5-4B checkpoint that accepts a local image path directly, letting the vision-language model score declared options without generating text. Ships scoring modes, a WebGPU browser demo and benchmark results measured on an RTX 3090.
Jev Models & Reimplementations#community#multimodal#open-models
Community
jiangxiluning/Visual-JevPython
jiangxiluning/Visual-JevPythonCloudflare's first open decision models on Workers AI: Clef, a 27B model that reads text, JSON, images, or video, and Clef-flash, a 9B model for the latency-sensitive hop, each returning a probability per option.
Jev Models & Reimplementations#community#cloudflare#decision-model
Community
The Hugging Face hub for Laya's open System 1 checkpoints: the 421M English model at the repo root, a 322M multilingual build in a subfolder, and the typed-decisions fine-tune, all Apache-2.0.
Jev Models & Reimplementations#community#open-models#local
Community
From the Jev guides
Original write-ups on what this entry is for.
Jev takes text only; Clef-omni adds image, audio, and video input and Liquid d1 adds images, each keeping the same typed-answer shape. Where that matters, where it does not, and the costs the vendors underplay.
Jev Models & Reimplementations#multimodal#open-models#decision-model
Read article
One model in September, a dozen by October. Clef, Liquid d1, Perplexity, and OpenAI all ship the same typed-answer shape now. How to read the rival benchmarks and pick on your own task.
Jev Models & Reimplementations#decision-model#benchmarks#pricing
Read article
Hundreds of repositories carry the system-one tag, but the ecosystem is a dozen ideas in different sizes. A tour of the families, the licenses, and what a download does not include.
Jev Models & Reimplementations#open-models#hugging-face#local
Read article
From the Jev community
Posts from builders shipping with Jev right now.