waddie/mini-1.0

TEXT GENERATIONConcurrency Cost:1Model Size:0.5BQuant:BF16Ctx Length:32kPublished:Apr 27, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

waddie/mini-1.0 is a 0.5 billion parameter instruction-tuned model, fine-tuned from Qwen2.5-0.5B-Instruct by Edward Fazackerley. It is specifically designed to emulate a casual, lowercase-heavy, and technical "random guy" persona, trained on AI Leaks community conversations. This model excels in generating human-like, informal dialogue for Discord bots or roleplay scenarios, with a context length of 32768 tokens.

Loading preview...

waddie/mini-1.0: The "Random Guy" Persona Model

waddie/mini-1.0 is a unique 0.5 billion parameter language model, fine-tuned by Edward Fazackerley from Qwen2.5-0.5B-Instruct. Unlike typical helpful AI assistants, this model is engineered to adopt a specific, informal persona, characterized by casual language, technical jargon, and a consistent use of lowercase.

Key Capabilities

  • Distinct Persona Emulation: Trained on curated conversation pairs from an AI Leaks community, it captures a "random guy" vibe, including slang and a slightly secretive tone.
  • Informal Communication: Generates responses in a casual, lowercase-only style, mimicking human-like interaction rather than formal AI.
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for more extended and coherent informal dialogues.

Good For

  • Discord Bots: Ideal for creating bots that interact with users in a casual, human-like manner.
  • Roleplay Scenarios: Excellent for applications requiring a specific, informal character voice.
  • Mimicking Insider Jargon: Suited for generating text that includes technical slang and an "insider" feel.

To achieve the best results, users are encouraged to prompt the model using all lowercase and minimal formal punctuation, following the recommended ChatML format.