Will the local GGUF fallback match the Jev API?

No. The README says the local model approximates Jev-style decisions from next-token probabilities in one forward pass, sends nothing over the network, and produces answers and probabilities that differ from Jev.

Where does my TypeSafe API key live for the Jev node?

Install typesafe-sdk in the ComfyUI Python environment and put the key as one line in api_key.txt inside the node folder. It is read on every run, ignored by Git, and only the Jev node contacts the API.

How do batched questions and the call cap work?

Questions queue while the script runs and send one request per state when a value is first used. The max_jev_calls default of 8 limits requests rather than questions, and identical questions ask once.

Why did my calculated float lose its fraction?

Numeric outputs are INT and fractional parts truncate, so handle rounding in script logic and use the centi node when FLOAT behavior is needed. Unassigned outputs default to None.

How do I install ScriptFlow, and what license is it under?

Clone the repository into ComfyUI/custom_nodes and restart ComfyUI. The project is GPL-3.0, and the plain MultiOutputScript and centi nodes need no extra packages.

How safe is the script interpreter?

Scripts run in an AST interpreter with restricted syntax and no file or OS access, and only the Jev node's jev functions reach a model or the TypeSafe API.

What is the default Jev call cap, and what counts against it?

max_jev_calls defaults to 8 requests per run. Questions about the same state share one request, so the cap limits requests rather than questions, and identical questions are asked only once.

CommunityJev Models & Reimplementations

ComfyUI-ScriptFlow

ComfyUI script nodes offering a restricted Python-like interpreter with multiple text and numeric ports, plus a Jev variant that branches on yes, choice, and score answers.

See highlights

Open github.com

Stars: 2
Language: Python
Published by: Community
Author: kantan-kanto
Last verified: Oct 2026

kantan-kanto/ComfyUI-ScriptFlow

User repository on GitHub

View on GitHub

Safe Python-like script node with a System One decision model: ask yes/no, choice, and score questions via TypeSafe AI's Jev API, with local GGUF LLMs as a fallback. Multiple inputs and outputs let one node handle math, logic, text editing, conditional routing, resolution selection, filename generation, and more.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

About this project

Safe Python-like script node for ComfyUI with a System One decision model. Ask yes/no, choice, and score questions via TypeSafe AI's Jev API, with local GGUF LLMs as a fallback, and use the answer directly in an if statement. MultiOutputScript (Jev) picks an aspect ratio for a text prompt through the TypeSafe API, lists every ratio's probability, and passes the width and height to an image node. A simplified version of the demo: connect a text prompt to intext1, and the answer picks the width and height.

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Routing Jev for Scoring

Used inJev for Software & DevTools

Languages: Python

Highlights

Calls jev.yes, jev.noul, jev.choice, jev.probabilities, and jev.score directly in if statements with probabilities available to scripts.
Uses the TypeSafe API as primary and an instruction-tuned GGUF with a chat template as an offline fallback using the same scripts.
Runs scripts in a restricted AST interpreter with no file or OS access; only the Jev functions reach a model or API.
Ships a default script that picks one of eight aspect ratios, sizes to the requested megapixels, and lists every ratio probability.

Watch out

The local GGUF fallback only approximates Jev-style decisions and its answers and probabilities differ; numeric outputs are INT with truncation. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

jevons

Embeddable System One server matching TypeSafe HTTP shape, scoring LFM2.5-1.2B logits in code with optional Laya and cloud failover.

Jev Models & Reimplementations#community#github

Community

gopalanj/jevonsPython

Bespoke Nimble

2.2k

Open recipe for a local Jev-style decision model: a Qwen3.5-9B LoRA that scores one allowed answer token per Choice, boolean or rubric Score question, with data pipeline, training config and eval harness. Reports 90.1% on its 324-example holdout against Jev's 93.2%.

Jev Models & Reimplementations#community#lora#qwen

Community

bespokelabsai/nimblePython

Julia 1

Supersonic Labs' 144.3M-parameter decision model on mmBERT-small: choice, score, and noul answers from supplied options in one pass, 73.15% on 2,000 typed decisions, Apache-2.0.

Jev Models & Reimplementations#community#open-models#local

Community

Liquid AI d1-3B

Liquid AI's 3.12B decision model built on LFM2.5-VL: state plus typed questions in one forward pass with zero output tokens, 48.57 on Decision Index 0.2.1 and 8 ms per decision on an RTX 4090.

Jev Models & Reimplementations#community#open-models#local

Community

Decider Chat Gemma-4-31B

An untrained readout: stock Gemma-4-31B weights plus a decider config that reads option letters at the answer slot, scored #2 of 70 on Decision Index v0.2.1 at 57.33 against Jev 1.13.0's 57.91.

Jev Models & Reimplementations#community#open-models#local

Community

Drex DLM

Nace.AI's 8B decision model on NVIDIA's Efficient-DLM diffusion backbone: a 256-dim pointer head scores every option in one pass, behind the same POST /v1/systemone API as its hosted service.

Jev Models & Reimplementations#community#open-models#local

Community

From the Jev guides

Original write-ups on what this entry is for.

Multimodal Decision Models: What Clef and d1 Change

Oct 12, 2026

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

The Decision-Model Price War

Oct 11, 2026

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

What Is Actually on Hugging Face

Oct 8, 2026

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

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai