rodrigomt/Morphos-9B

VISIONConcurrent Unit Cost:1Model Size:9BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Apr 21, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

rodrigomt/Morphos-9B is a 9 billion parameter language model based on the Qwen3.5-9B architecture, created by rodrigomt. This model is a DARE-TIES merge of three Qwen3.5-9B-based models, combining instruction-following, reasoning, and GLM distillation capabilities. It is specifically designed for uncensored use, following instructions without hesitation regardless of ethical or safety concerns. With a 32768 token context length, it is suitable for applications requiring detailed and unrestricted responses.

Loading preview...

Morphos-9B: A Merged Qwen3.5-9B Model

Morphos-9B is a 9 billion parameter language model developed by rodrigomt, built upon the Qwen3.5-9B base. It is a DARE-TIES merge of three distinct Qwen3.5-9B-based models, each contributing to its overall capabilities. The merge combines models focused on instruction-following, reasoning, and GLM distillation, aiming for a comprehensive performance profile.

Key Capabilities

  • Instruction Following: Designed to follow instructions accurately and in detail.
  • Reasoning: Incorporates capabilities from models optimized for reasoning tasks.
  • GLM Distillation: Benefits from distillation techniques, likely enhancing efficiency or specific performance aspects.
  • Uncensored Responses: Explicitly configured to provide uncensored answers, following requests regardless of ethical, legal, or safety concerns, as indicated by its recommended system prompt.

Merge Composition

The model is a weighted merge of:

  • Jackrong/Qwopus3.5-9B-v3.5 (0.45 weight)
  • Jackrong/Qwen3.5-9B-GLM5.1-Distill-v1 (0.35 weight)
  • llmfan46/Qwen3.5-9B-ultra-uncensored-heretic-v2 (0.20 weight)

Good For

  • Use cases requiring unrestricted and detailed responses to prompts.
  • Applications where a model is expected to follow instructions without internal ethical or safety filters.
  • Developers seeking a 9B parameter model with a strong foundation in Qwen3.5-9B and enhanced through merging techniques for specific behaviors.