Xx-Vexento-xX/roast-bot-qwen-1b

TEXT GENERATIONConcurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 26, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The Xx-Vexento-xX/roast-bot-qwen-1b is a 1.5 billion parameter Qwen 2.5 model fine-tuned by Xx-Vexento-xX with a 32768 token context length. This model specializes in generating witty comebacks, absurdist tech comparisons, and casual language, mimicking a specific texting style. It is optimized for conversational AI applications requiring a distinct, humorous, and aggressive response style.

Loading preview...

Overview

This model, Xx-Vexento-xX/roast-bot-qwen-1b, is a fine-tuned version of the Qwen 2.5 1.5B Instruct model, specifically adapted to emulate a unique texting style characterized by witty comebacks, absurdist tech comparisons, and casual language. It was trained using LoRA (r=64, alpha=128) on 90 conversation examples, augmented to 270, covering roasts, casual chat, and identity responses.

Key Capabilities

  • Distinct Personality: Responds in a specific, humorous, and aggressive style, ideal for engaging and entertaining interactions.
  • Contextual Roasting: Capable of generating clever roasts in response to aggressive messages.
  • Casual Conversation: Handles casual messages with a laid-back tone.
  • Identity Awareness: Can accurately identify itself as an AI trained on a specific style when asked.

Good For

  • Chatbots with Personality: Developing conversational agents that require a unique, witty, and slightly aggressive persona.
  • Entertainment Applications: Creating interactive experiences where humorous and unexpected responses are desired.
  • WhatsApp Integrations: Designed with a potential integration into a WhatsApp bot, as detailed in its GitHub repository.