kepom/Qwythos-9B-Claude-Mythos-5-1M

VISIONConcurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 4, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Qwythos-9B is a 9 billion parameter reasoning model developed by Empero, built on a deeply uncensored Qwen3.5-9B base. It is post-trained on over 500 million tokens of high-quality Claude Mythos and Fable traces, featuring a 1,048,576-token context window. This model excels at complex reasoning tasks, particularly in cybersecurity, biomedical, and quantitative domains, and supports native function calling for tool use.

Loading preview...

Qwythos-9B: A Powerful 9B Reasoning Model

Qwythos-9B, developed by Empero, is a 9 billion parameter reasoning model based on a deeply uncensored Qwen3.5-9B architecture. It has been extensively post-trained on over 500 million tokens of high-quality Claude Mythos and Fable traces, incorporating chain-of-thought generated by Empero AI's internal tool, rethink. This training approach results in a compact, fast, and highly capable model.

Key Capabilities

  • 1,048,576-token context window: Features YaRN rope-scaling for a full 1M-token context, making it suitable for whole-codebase reasoning, multi-document research, and long agentic trajectories.
  • Enhanced Reasoning Performance: Demonstrates significant improvements over its base model, with +34 pts on MMLU and +30 pts on gsm8k-strict under matched evaluation conditions.
  • Native Function Calling: Supports OpenAI/Qwen3.5-style function calling out of the box, enabling seamless integration with tools like Python executors and web search.
  • Self-Correction with Tools: Proven to self-correct and provide source-cited, factually correct answers on challenging prompts when paired with appropriate tools.
  • Uncensored Design: Intentionally uncensored to engage seriously with technically demanding questions in domains like cybersecurity, red-teaming, biology, pharmacology, and clinical medicine, where over-aligned models often refuse or provide disclaimers.

Good For

  • Complex Reasoning Tasks: Excels in domains requiring deep analytical thought, such as cybersecurity, biomedical analysis, and quantitative problem-solving.
  • Long-Context Applications: Ideal for scenarios needing extensive context, including analyzing large codebases, synthesizing information from multiple documents, and managing long agentic interactions.
  • Tool-Augmented Workflows: Designed for deployment in retrieval-augmented agentic settings where models verify specifics and integrate external information.
  • Technical and Scientific Research: Provides substantive engagement with specialized technical questions without refusal or boilerplate disclaimers.

Finetunes
1 models