eugenehp/qwen3-0.6b

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.8BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 23, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The eugenehp/qwen3-0.6b model is a 0.8 billion parameter language model based on the Qwen3 architecture, developed by Qwen and redistributed by eugenehp for RLX staging. This model is designed for efficient deployment within the RLX ecosystem, offering a compact yet capable solution for various language processing tasks. It provides a stable fetch target for RLX recipes, making it suitable for developers integrating Qwen3 capabilities into their RLX-based applications.

Loading preview...

Overview

This repository, eugenehp/qwen3-0.6b, provides a staging redistribution of the Qwen3-0.6B model, originally developed by Qwen. It is specifically packaged for use with the RLX framework, offering a stable and accessible checkpoint for RLX runners. The model has 0.8 billion parameters and supports a context length of 32768 tokens, making it a compact yet capable option for various language tasks within the RLX ecosystem.

Key Features

  • RLX Integration: Optimized for deployment and execution within the RLX framework, with a dedicated rlx-qwen3 crate.
  • Compact Size: At 0.8 billion parameters, it offers a balance between performance and resource efficiency.
  • Standard File Formats: Includes model.safetensors (1.4 GiB) and a quantized Qwen3-0.6B-Q4_K_M.gguf (378.3 MiB) for flexible deployment options.
  • Comprehensive Tokenization: Comes with tokenizer.json, vocab.json, and merges.txt for consistent text processing.

Usage and Licensing

Developers can quickly get started by downloading the model and running it with the provided rlx-qwen3 crate, supporting devices like metal. The model inherits the Apache License 2.0 from its upstream source, Qwen/Qwen3-0.6B, ensuring clear terms for redistribution and use. This specific Hub repository serves as a convenient fetch target for RLX projects, while the original upstream checkpoint is recommended for general use outside of RLX staging.

Finetunes
2 models