Jev directory
CommunityJev Models & Reimplementations
MiniCPM-OCR-Jev-mxfp8
Hugging Face model repository for MiniCPM-OCR-Jev-mxfp8, published by the nativ-community organization in mxfp8 precision. Its chat template handles image and video inputs, XML-style tool calls, and reasoning content.
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- Hosted on Hugging Face under the nativ-community organization.
- Model name indicates mxfp8 precision for MiniCPM-OCR.
- Chat template renders image and video content through vision tokens.
- Tool calls are emitted as XML inside tool_call tags.
- Pad token is <|endoftext|> with unk_token set to null.
Questions about this entry
Short answers drawn from the linked project or page.
It is a Hugging Face model repository published by the nativ-community organization. The name identifies it as a MiniCPM-OCR variant in mxfp8 precision, and the repository ships a chat template that handles text, images, video, and tool calls.
The repository sits under the nativ-community organization on Hugging Face, at the path nativ-community/MiniCPM-OCR-Jev-mxfp8. The digest does not name individual maintainers behind the account.
The template renders text, image, and video content items. Images become <|vision_start|><|image_pad|><|vision_end|> blocks and videos become <|vision_start|><|video_pad|><|vision_end|> blocks. System messages are explicitly blocked from containing images or videos.
Tool calls are written as XML: a <tool_call> block containing a <function=name> element with <parameter=...>value</parameter> entries. The template instructs that required parameters must be specified and that optional reasoning may precede the call but not follow it.
Assistant messages may carry a reasoning_content field, or the template will split content on <think> and </think> tags. Reasoning is rendered inside think tags only for turns after the last user query; earlier assistant turns are rendered without it.
The template metadata lists pad_token as <|endoftext|> and unk_token as null. The digest does not list additional special tokens or vocabulary details.
The provided digest does not include license text or usage terms for this repository. Check the model card on Hugging Face directly before redistributing or deploying the weights.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Standard One SH (StandardThinking)
LinkedIn: What multimodal decision models actually change, and what they do not.
LinkedInJevDirectory.org
Hacker News: Decision Index 0.2.1 scores open reproductions of Jev on a frozen suite.
jackbravo4 pointsDecision Index 0.2.1 – 70 open reproductions of TypeSafe Jev's Decision ModelRead the thread on Hacker NewsHacker News: Pointer to a DiffusionGemma plus vLLM tutorial for a Jev-shaped classifier.
1 pointHow to build a Jev-style classifier with DiffusionGemma and vLLMRead the thread on Hacker NewsMore Jev resources like this
Open Jev-style decision model built on LiquidAI's LFM2.5-350M backbone, a clone of kev. Serves the System One API contract so the TypeSafe SDK works against a local server, returning probabilities for typed questions from one forward pass.
Jev Models & Reimplementations#open-models#python#local
Communityfranckverrot/lev
Satirical Python script that answers a query with DO or DO_NOT plus a random confidence percentage, presented as a natural alternative to Jev. MIT-licensed, Python 3.8+, one dependency, no API keys or model weights.
Jev Models & Reimplementations#github#python#open-models
CommunityAhmad170412/Koin
Rust decision engine that runs TypeSafe AI's Jev protocol locally: one ONNX forward pass answers noul, choice, and score questions with calibrated probabilities and confidence. Serves a dashboard and the /v1/systemone API on 127.0.0.1:8000.
Jev Models & Reimplementations#rust#local#open-models
Communityalnyx-dev/jev-rs
A native MLX runtime for Laya's open typed-decision weights: short decisions on Apple Silicon, with no generated tokens, PyTorch, or cloud API.
Jev Models & Reimplementations#community#python#open-models
Community
mizorewww/laya-mlxPython
mizorewww/laya-mlxPythonAn open-model semantic-if engine positioned as a local Jev-style decision stack, separate from the original SemIf repo but sharing the same author.
Jev Models & Reimplementations#community#python#open-models
Community
TheoLeeCJ/SemIf-OpenJevPython
TheoLeeCJ/SemIf-OpenJevPythonLaya typed decisions through Apple Core ML and the Neural Engine, with the author's speed and energy benchmarks for short decisions on an M3 Max.
Jev Models & Reimplementations#community#python#open-models
Communitymizorewww/laya-coremlPython
mizorewww/laya-coremlPythonFrom 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.