AIRAWAT, a flagship initiative of the Ministry of Electronics and Information Technology (MeitY), Government of India, led by C-DAC Pune, has been envisioned to realize the Hon’ble Prime Minister Shri Narendra Modi’s vision of “AI for All”—to make Artificial Intelligence in India, and Artificial Intelligence work for India.

The 200 AI Petaflops Proof-of-Concept AI Research Analytics and Knowledge Dissemination Platform (AIRAWAT) is built on a scalable infrastructure to act as a common computational cloud platform for AI. The system has been ranked 75th globally. It was declared so in the 61st edition of the Top 500 Global Supercomputing List at the International Supercomputing Conference (ISC 2023) in Germany. It places India at the forefront of AI Supercomputing nations worldwide.

It provides dense GPU compute resources specifically designed for AI workloads and is currently being used by over 65 startups, MSMEs, PSUs, industries, and premier academic and research institutions across the country. To incentivize the Indian startup ecosystem, AIRAWAT offers free credits worth 200 GPU hours along with 1 TB of high-throughput storage at no cost.

Common Computational AI Cloud

A national AI computing platform connecting research centers, academia, startups, and industry via the National Knowledge Network.

Empowering Indigenous AI

Enables development of indigenous AI-enabled models, products and solutions aligned with Atmanirbhar Bharat an initiative of the Government of India.

Solving Real-World Challenges

Accelerates AI breakthroughs in:

• NLP: Indian languages, speech & translation
• Agriculture: Smart farming & supply chains
• Healthcare: Diagnostics, telemedicine, drug discovery
• Security: Surveillance, anomaly detection
• Finance: Fraud detection, risk analytics
• Education: Personalized learning & content
• Defence & Automotive: Smart mobility, predictive maintenance

Collaborative Ecosystem

Brings researchers, industry, and startups together to innovate and share AI resources.

Scalable & Affordable

Provides high-performance, cost-effective AI compute for rapid prototyping and deployment.

AI Training & Deployment at Scale

Model Training

Run large-scale training and fine-tuning with a powerful batch scheduler, GPU-optimized environment, and containerized workflows for speed and reproducibility.

Model Hosting

Powerful Pre-Trained & Domain-Specific AI Models

Cutting-Edge Foundation Models

Access a wide range of state-of-the-art GenAI models, including:

• Llama: 3.2 11B Vision Instruct, 3.3 70B Instruct, 4 Scout Instruct
• GPT-OSS: Open-weight 20B \& 120B
• NVIDIA Nemotron: Nano 12B v2
• Mistral: Small 24B Instruct (2501)
• Magistral: Small (2506)
• Krutrim8

Specialized AI Capabilities

• Text Processing & Ranking: Jina Reranker v2 Base Multilingual
• OCR: Nanonets/Nanonets-OCR-s
• Speech Recognition: Whisper Large v3 & Multi-ASR Offline-ASR BLS Ensemble
• Indic Language Translation: English->Gujarati, Hindi, Marathi

Flexible Deployment

Consume models via dedicated Kubernetes clusters or fully managed API endpoints, tailored to your workload and scalability needs.

Future-Ready AI Cloud

The upcoming AIRAWAT AI Cloud will offer enterprise-grade GPU and CPU resources, supporting VMs, containers, and bare metal, along with auto-scaling, quota management, and high availability.

National PARAM Supercomputing Facility (NPSF)

Centre for Development of Advanced Computing

C-DAC Innovation Park, Panchawati,

Pashan, Pune - 411 008 INDIA

E-mail

airawat-outreach[at]cdac[dot]in