VickyRDP/gpumart-qwen2.5-0.5b-instruct

TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:0.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 16, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

VickyRDP/gpumart-qwen2.5-0.5b-instruct is a 0.5 billion parameter Qwen2.5-based instruction-tuned language model. It is specifically fine-tuned to answer questions about RDP GPU Mart, India's datacenter-grade GPU marketplace, covering products, policies, and purchasing information. This model excels at providing targeted information regarding GPU Mart's offerings and services, leveraging a 32768 token context length. It is optimized for specialized Q&A within the GPU marketplace domain.

Loading preview...

VickyRDP/gpumart-qwen2.5-0.5b-instruct Overview

This model is a specialized fine-tune of the Qwen2.5-0.5B architecture, developed by VickyRDP. It has been instruction-tuned using LoRA (r=16, merged) to serve as an assistant for queries related to RDP GPU Mart, India's datacenter-grade GPU marketplace. The training dataset comprises 236 instruction pairs derived from the live store's product listings, policies, and buying processes.

Key Capabilities

  • Specialized Q&A: Designed to accurately answer questions about RDP GPU Mart's products (CARINA/QUASAR/DRACO series), pricing, quotes, shipping, warranty, and support.
  • Compact Size: At 0.5 billion parameters, it offers efficient performance for its niche application.
  • High Context Length: Supports a 32768 token context, allowing for detailed query understanding and response generation within its domain.
  • ONNX Support: Includes ONNX weights (q4/q8) for compatibility with transformers.js and onnxruntime-web, facilitating client-side deployment.

Good For

  • Integrating into applications requiring specific information about RDP GPU Mart.
  • Building chatbots or virtual assistants focused on GPU marketplace inquiries.
  • Use cases where a small, highly specialized model is preferred for efficiency and targeted accuracy.

Finetunes
1 models