Dhruv-27B Open-Source Language Model - Free Deployment to Boost Academic, Logical, and Factual Evaluation Reasoning

Dhruv 27B

Developed by vicharai

Dhruv-27B-preview is a 27 billion parameter language model based on Gemma3, fine-tuned for advanced reasoning tasks, suitable for academic, logical, and factual evaluation.

Large Language Model

Transformers

#27 billion parameter inference #Academic Q&A optimization #Multilingual factual reasoning

Downloads 18

Release Time : 3/18/2025

Model Overview

Dhruv-27B-preview is a high-performance language model optimized for reasoning, Q&A, and factual accuracy, suitable for research, enterprise applications, and general knowledge understanding.

Model Features

Advanced reasoning capabilities

Specially optimized for reasoning tasks, capable of deep contextual reasoning and chain-of-thought generation.

High-performance benchmarking

Excellent performance in benchmarks such as MMLU and GPQA, demonstrating strong knowledge understanding and reasoning abilities.

Efficient inference

Supports int8 quantization, improving inference efficiency.

Model Capabilities

Advanced reasoning

Q&A tasks

Factual evaluation

Multilingual processing

Chain-of-thought generation

Use Cases

Academic research

Academic Q&A

Used to answer complex academic questions, providing accurate explanations and analysis.

Achieved 80.0% accuracy in MMLU 5-shot testing.

Enterprise applications

Fact-checking

Used for fact-checking and information verification in enterprise environments.

AI agents

General reasoning agent

Serves as a core component for AI agents requiring strong knowledge understanding.

🚀 Model Card for Dhruv-27b-preview

Dhruv-27b-preview is a powerful language model based on Gemma3, fine - tuned for advanced reasoning tasks, suitable for various research and enterprise applications.

🚀 Quick Start

To access Gemma on Hugging Face, you’re required to review and agree to Google’s usage license. To do this, please ensure you’re logged in to Hugging Face and click the "Acknowledge license" button below. Requests are processed immediately.

✨ Features

Dhruv-27B-preview is a 27 - billion parameter language model based on Gemma3, fine - tuned for advanced reasoning tasks.
It is part of the Gemini Reasoning Series, optimized for high performance on academic, logical, and factual evaluations.
Capable of deep contextual reasoning and chain - of - thought generation, suitable for research, enterprise applications, and AI agents requiring robust general knowledge understanding.

📚 Documentation

Key Metrics

Dhruv-27b-preview achieves strong performance on key benchmarks:

+------------------------+--------+
|        Metric         | Score  |
+------------------------+--------+
| MMLU (5-shot)          | 80.0%  |
| GPQA (0-shot)          | 52.0%  |
+------------------------+--------+

Model Architecture

Base Model: Gemma3
Model Size: 27B parameters
Type: Decoder-only Transformer (causal LM)
Precision: bf16 with int8 quantization for inference efficiency
Training Objective: Instruction-tuned with emphasis on reasoning, question answering, and factual correctness

Intended Use

Research and academic QA tasks
General-purpose reasoning agents
Multilingual factual reasoning
Enterprise AI tools requiring high factual accuracy and depth

USAGE

Use a System Template

Limitations

Not optimized for creative generation or dialog tasks.
May hallucinate in areas with limited training data.
Requires fine-grained prompt engineering for complex instructions.

Citation

@misc{vicharai_dhruv27b,
  title={Dhruv-27B: Gemma3 Reasoning Model},
  author={vicharai},
  year={2025},
  publisher={vicharai},
  url={https://vichar.io}
}

Developed by vichar ai Huggingface Website

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご