empero-ai/Qwythos-9B-Claude-Mythos-5-1M

Hugging Face
VISIONConcurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 19, 2026License:apache-2.0Architecture:Transformer0.8K Open Weights Featherless Exclusive Warm

Qwythos-9B-Claude-Mythos-5-1M by Empero is a 9 billion parameter reasoning model built on a Qwen3.5-9B base, post-trained on over 500 million tokens of Claude Mythos and Fable traces. It features a 1M-token context window via YaRN rope-scaling, native function calling, and self-correction with tools. This model excels in technically demanding domains like cybersecurity, biomedical analysis, and quantitative reasoning, offering significant performance gains over its base model on benchmarks like MMLU and GSM8K.

Loading preview...

Qwythos-9B-Claude-Mythos-5-1M: A Specialized Reasoning Model

Empero's Qwythos-9B is a 9 billion parameter model based on Qwen3.5-9B, fine-tuned with over 500 million tokens of high-quality Claude Mythos and Fable traces. It is designed for advanced reasoning tasks, particularly in technical and scientific fields.

Key Capabilities & Features

  • 1M-Token Context Window: Utilizes YaRN rope-scaling for a 1,048,576-token context, enabling whole-codebase reasoning, multi-document research, and long agentic trajectories.
  • Enhanced Reasoning: Achieves significant improvements over the base Qwen3.5-9B, with +34 pts on MMLU and +30 pts on GSM8K-strict.
  • Native Function Calling: Supports OpenAI/Qwen3.5-style function calling out-of-the-box, allowing for tool use and self-correction.
  • Tool-Augmented Reliability: Demonstrated success in self-correcting and providing source-cited, factually correct answers across complex prompts using Python executors and web search.
  • Uncensored: Intentionally uncensored to engage substantively with technically demanding questions in cybersecurity, red-teaming, biology, pharmacology, and clinical medicine.

Ideal Use Cases

  • Technical & Scientific Research: Analyzing multi-paper corpora, complex biomedical questions, and pharmacological reasoning.
  • Code & Software Development: Whole-codebase reasoning, cross-file refactoring, and defect finding.
  • Agentic Workflows: Deployments requiring tool use, self-correction, and retrieval-augmented generation.
  • Cybersecurity & Red-Teaming: Detailed analysis and explanations of complex security concepts and methodologies.

Finetunes
4 models