Jev in 25 lines of Python

NobodyWho published a parody blog post on September 22, 2026 showing a Jev-style decision model implemented in 25 lines of Python using the Qwen3-0.6B-GGUF model via llama-cpp-python, which classifies an email prompt into Legitimate, Spam, or Phishing with probabilities of 0.031, 0.084, and 0.885 respectively. Author Duarte O.Carmo argues the approach is fast, local, and avoids sending data to an API, and links to fuller open implementations OpenJev, openjev-sglang, and OpenJev on DiffusionGemma.

Jev in 25 lines of Python Everyone and their mom is talking about Jev. Jev this, Jev that. Everyone on Twitter is all over Jev, how it's the next frontier of large language models and the AI paradigm. We don’t really think so. So here's Jev in 25 lines of Python. Load the model. /// script requires-python = " =3.12" dependencies = "huggingface-hub", "llama-cpp-python", "numpy" /// import numpy from llama cpp import Llama Really, you can use any GGUF model from https://huggingface.co/models?library=gguf model = Llama.from pretrained repo id="Qwen/Qwen3-0.6B-GGUF", filename="Qwen3-0.6B-Q8 0.gguf", n ctx=512, logits all=True, verbose=False, Load the prompt and define your choices. labels = "A", "B", "C" choices = "Legitimate", "Spam", "Phishing" email = "Payroll asks for your password on a non-company sign-in page." options = "\n".join f"{label}. {choice}" for label, choice in zip labels, choices, strict=True prompt = f"""<|im start| system Choose one option.<|im end| <|im start| user Email: {email}\n\n{options}<|im end| <|im start| assistant <think \n\n</think \n\n""" model.eval tokens=model.tokenize text=prompt.encode , add bos=False, special=True Massage the logits into probabilities. logits = model.scores model.n tokens - 1 token ids = model.tokenize text=label.encode , add bos=False 0 for label in labels choice logits = numpy.asarray logits token id for token id in token ids logprobs = choice logits - numpy.logaddexp.reduce choice logits probabilities = numpy.exp logprobs for name, scores in "Logits", choice logits , "Log probabilities", logprobs , "Probabilities", probabilities , : values = numpy.round scores.astype float , 3 .tolist print f"{name}:", dict zip choices, values, strict=True Logits: {'Legitimate': 26.254, 'Spam': 27.262, 'Phishing': 29.614} Log probabilities: {'Legitimate': -3.482, 'Spam': -2.474, 'Phishing': -0.122} Probabilities: {'Legitimate': 0.031, 'Spam': 0.084, 'Phishing': 0.885} There. That’s Jev. But no, you don’t understand Jev Yeah, we know. - We don't call it a System One decision model. - We didn’t call an API. - We didn't create a bunch of synthetic data. - We didn't train a model with Reinforcement Learning for Calibrated Decisions RLCD to calibrate the decisions and probabilities even though they are not always correct . But yes. This is Jev. - It classifies: it gets a prompt with choices and outputs probabilities. - It's fast. - It's local. - You don't send your data anywhere else. And we like not sending your data anywhere else. Check out NobodyWho https://github.com/nobodywho-ooo/nobodywho . note: this is a parody blog post, see these links for better/more complete open implementations of Jev: OpenJev, openjev-sglang, and OpenJev on DiffusionGemma. Everything NobodyWho do is open-source, please leave a star on Github https://github.com/nobodywho-ooo/nobodywho to support us ❤️ Published Sep 22, 2026 by Duarte O.Carmo