Fox-AI-by-teolm30/fox-1.6

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:2BQuant:BF16Context Size:32kTool Calling:SupportedPublished:May 28, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Fox 1.6 is a compact, 1.7 billion parameter language model developed by teolm30, forked from Qwen/Qwen3-1.7B. This model is designed for general-purpose text generation, focusing on on-device or low-footprint inference. It is optimized for deployment on consumer hardware, making it suitable for assistant-style chat and completion tasks.

Loading preview...

Fox 1.6: A Compact Qwen3 Fork

Fox 1.6 is a lightweight, 1.7 billion parameter language model developed by teolm30, based on the Qwen/Qwen3-1.7B checkpoint. This model is specifically designed to be easily deployable on consumer hardware, offering a more accessible option compared to much larger LLMs.

Key Characteristics

  • Compact and Device-Friendly: Optimized for on-device or low-footprint inference environments.
  • General-Purpose Text Generation: Capable of handling a variety of text generation tasks.
  • Forked from Qwen3-1.7B: Leverages the architecture and initial training of the Qwen3 1.7B model.

Intended Use Cases

  • On-Device Inference: Ideal for applications requiring local processing without extensive computational resources.
  • Assistant-Style Chat: Suitable for building conversational agents and chatbots.
  • Text Completion: Can be used for generating continuations of given text prompts.
  • Rapid Experimentation: Provides a compact base model for quick prototyping and development.

To evolve Fox 1.6 into a genuinely new model, the recommended next step is fine-tuning on a curated instruction dataset and subsequent evaluation against specific benchmarks.

Finetunes
2 models