zenlm/zen-eco

TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Nov 13, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

zen-eco is an efficient 0.8 billion parameter general-purpose language model from the Zen Eco line, repackaged from Alibaba Qwen's Qwen3-0.6B architecture. This permissively-licensed model is designed for broad applications, offering a compact yet capable foundation for various NLP tasks. With a 32768 token context length, it provides a solid base for developers seeking an OSS-clean, general-purpose LLM.

Loading preview...

zen-eco: An Efficient General-Purpose Language Model

zen-eco is a compact yet capable 0.8 billion parameter language model, part of the Zen Eco line. It is a repackaged and permissively-licensed redistribution of Alibaba Qwen's Qwen3-0.6B model, utilizing the Qwen3 architecture (Qwen3ForCausalLM). This model is not trained from scratch but rather provides an OSS-clean version of the established Qwen3-0.6B.

Key Capabilities

  • Efficient Performance: Designed for general-purpose language tasks with a small parameter count (0.8B).
  • Broad Applicability: Suitable for a wide range of NLP applications due to its foundation model nature.
  • Extended Context: Features a notable context length of 32768 tokens, allowing for processing longer inputs.
  • Permissive Licensing: Distributed under the apache-2.0 license, making it suitable for commercial and open-source projects.

Good For

  • Developers seeking a compact, general-purpose LLM for resource-constrained environments.
  • Applications requiring a permissively-licensed foundation model.
  • Prototyping and deployment where efficiency and a solid base model are priorities.