prithivMLmods/Nenque-MoT-0.6B-Elite14

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 1, 2025License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Loading

prithivMLmods/Nenque-MoT-0.6B-Elite14 is a compact 0.6 billion parameter model, fine-tuned from Qwen3-0.6B using the Mixture of Thoughts (MoT) dataset with a focus on math expert clusters. This model excels in mathematical reasoning, code generation, and structured technical inference across STEM and multilingual technical domains. It is optimized for low-resource environments, edge devices, and GPUs with limited VRAM, delivering elite-level precision despite its small size. Its primary strength lies in generating structured outputs like Markdown, JSON, and LaTeX for technical documentation and data.

Loading preview...

Nenque-MoT-0.6B-Elite14: Compact Technical Reasoning Model

Nenque-MoT-0.6B-Elite14 is a 0.6 billion parameter model, fine-tuned from Qwen3-0.6B by prithivMLmods. It leverages a specialized Mixture of Thoughts (MoT) dataset, with a strong emphasis on math expert clusters, to achieve high efficiency in symbolic tasks. This model is designed for robust performance in resource-constrained environments.

Key Capabilities

  • Elite Mathematical Reasoning: Excels in solving algebraic equations, calculus, and symbolic logic step-by-step, making it suitable for educational tools and STEM support.
  • Compact Code Assistant: Generates concise and explainable code in languages like Python and JavaScript, ideal for prototyping, code tutoring, and bug diagnosis.
  • Structured Output Generation: Supports output in formats such as Markdown, JSON, LaTeX, and tabular formats, which is valuable for technical documentation and data generation.
  • Multilingual Technical Mastery: Provides consistent results across over 20 languages for mathematical and coding tasks.
  • Lightweight Inference: Optimized for edge devices and GPUs with limited VRAM, enabling high-quality results on constrained systems.

Intended Use Cases

  • Step-by-step mathematical reasoning and symbolic computation.
  • Lightweight multilingual code generation and debugging.
  • Structured content generation (e.g., LaTeX, JSON, Markdown).
  • Academic tutoring and technical assistant roles.
  • Deployment in resource-constrained or edge scenarios.

Limitations

This model is not designed for extended creative generation or conversational fluency. Its context length may impact performance on very long multi-step tasks, and it may underperform on general chat or abstract logic tasks as it is specialized for technical domains and structured outputs.

Finetunes
2 models