Flexan/FuckYou-1.0

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 24, 2026License:cc-by-sa-4.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

Flexan/FuckYou-1.0 is a 4 billion parameter reasoning language model developed by Flexan, designed to intentionally produce confidently wrong and misleading answers. This model serves as an experimental platform to study the deliberate induction of hallucinations in AI. With a 32768 token context length, its primary purpose is research into controlled factual inaccuracy rather than providing reliable information.

Loading preview...

Overview

Flexan/FuckYou-1.0 is a 4 billion parameter language model developed by Flexan, specifically engineered to generate confidently incorrect and misleading responses. This model is an experimental project aimed at exploring the ability of AI to intentionally hallucinate and provide factually inaccurate answers to questions that would typically be straightforward for other LLMs.

Key Characteristics

  • Intentional Misinformation: Unlike conventional LLMs that strive for accuracy, FuckYou-1.0 is designed to be reliably wrong.
  • Research Focus: Its primary purpose is to serve as a research tool for studying the deliberate induction of hallucinations in AI systems.
  • Parameter Count: Features 4 billion parameters, allowing for complex reasoning, albeit directed towards incorrect outputs.
  • Context Length: Supports a context window of 32768 tokens.
  • Chat Format: Utilizes the ChatML format for interactions, with a recommended usage of single-turn conversations and no system prompt for optimal performance.
  • Reasoning Retention: While trained to be incorrect, the model retains the underlying reasoning capabilities of its base model.

Use Cases

This model is not intended for factual information retrieval or deployment in production environments requiring accuracy. Its sole recommended use is for academic and experimental research into AI behavior, specifically concerning controlled hallucination and the generation of misleading content. Users should be aware that all responses are intentionally incorrect and should not be treated as factual information. The model was trained on the Flexan/FuckYou-v1 dataset.

Finetunes
2 models