Thox-ai/ThoxEdge-0.5B

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 11, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Thox-ai/ThoxEdge-0.5B is a 0.5 billion parameter model derived from Qwen/Qwen2.5-0.5B-Instruct, fine-tuned by Thox-ai. This model is specifically optimized and quantized for on-device use within the ThoxMini tier. Its primary purpose is to function as an on-device assistant, excelling at chat and agent/tool-use routing for the THOX device program.

Loading preview...

ThoxEdge-0.5B: On-Device AI for ThoxMini

ThoxEdge-0.5B is a specialized 0.5 billion parameter language model developed by Thox-ai. It is built upon the Qwen/Qwen2.5-0.5B-Instruct architecture and has undergone fine-tuning using Thox-ai's proprietary chat and agent/tool-use corpora. A key differentiator for this model is its optimization and quantization specifically for the ThoxMini device tier, making it suitable for efficient on-device deployment.

Key Capabilities

  • On-device assistance: Designed to run efficiently on embedded hardware.
  • Chat functionality: Fine-tuned for conversational interactions.
  • Agent/Tool-use routing: Capable of understanding and directing tasks involving external tools or agents.

Intended Use Cases

  • Thox device program: Serves as the core AI for the ThoxMini ecosystem.
  • Edge AI applications: Ideal for scenarios requiring local processing and reduced latency.

Limitations

  • Not intended for safety-critical, medical, or legal decision-making without human oversight.

This model is currently in a 'skeleton' status, with weights pending final fine-tuning and quantization.

Finetunes
1 models