drvivekpoojary/Open-GVP-Qwen3-4B-Instruct

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 2, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

drvivekpoojary/Open-GVP-Qwen3-4B-Instruct is a 4 billion parameter Qwen3-based instruction-tuned language model developed by Dr. Vivek Poojary. It is specifically fine-tuned on approximately 15,000 high-quality question-answer pairs derived from European Medicines Agency (EMA) Good Pharmacovigilance Practices (GVP) guidelines. This model excels at regulatory interpretation and pharmacovigilance operations, making it ideal for GVP knowledge assistance and integration into Retrieval-Augmented Generation (RAG) pipelines for pharmacovigilance.

Loading preview...

Open-GVP-Qwen3-4B-Instruct Overview

Open-GVP-Qwen3-4B-Instruct is a specialized 4-billion parameter language model, domain-adapted from Qwen/Qwen3-4B-Instruct-2507. It has been fine-tuned using LoRA on a dataset of approximately 15,000 high-quality question-answer pairs derived from the official European Medicines Agency (EMA) Good Pharmacovigilance Practices (GVP) guidelines. This model is designed to provide expert assistance in pharmacovigilance, covering all GVP Modules and related Addendums.

Key Capabilities

  • Domain Specialization: Highly proficient in EMA GVP guidelines, including ICSR management, signal management, risk management systems, and periodic safety update reports.
  • Instruction-Tuned: Optimized for question-answering within the pharmacovigilance domain.
  • Flexible Deployment: Available in merged Safetensors for high-performance inference (Transformers, vLLM) and various GGUF quantized versions (BF16, Q8_0, Q6_K) for local and low-resource deployments (Ollama, llama.cpp).
  • RAG Integration: Performs best when integrated into a Retrieval-Augmented Generation (RAG) pipeline alongside official GVP PDF documents.

Recommended Use Cases

  • GVP Knowledge Assistant: Answering questions on GVP modules, definitions, and requirements.
  • PV Staff Training & Onboarding: Supporting the training of new pharmacovigilance team members.
  • RAG Pipeline Generator: Serving as the generative component in RAG systems with GVP documents.
  • Internal Regulatory Chatbot: Powering company-internal PV compliance assistants.
  • Offline/Edge Deployment: Suitable for air-gapped environments and devices with limited resources, especially with GGUF formats.

Finetunes
1 models