New Delhi, India - E2E Networks Ltd., a prominent provider of AI infrastructure and cloud services, has announced the global launch of its latest product, Blaze. This new service, launched on 2nd November 2026, is a pay-per-token inference service for open AI models, marking a significant expansion in E2E’s AI offerings.
Blaze is engineered on the NVIDIA DSX OS, giving developers and enterprises a single OpenAI-compatible API to access leading open models. The service is powered by NVIDIA GPU infrastructure, ensuring efficient production inference services hosted within India. Customers will only be charged for the tokens they use, making Blaze an economically viable option for businesses transitioning AI models from pilots to production.
“Blaze is where you can run your production inference,” said Tarun Dua, Founder and Managing Director of E2E Networks. “There is customer demand for frontier-class open weight models running on an NVIDIA Cloud Partner (NCP) with NVIDIA reference design clusters as Sovereign Compute. Blaze provides this capability on E2E Cloud in India.”
The product caters to both domestic and international markets, reflecting a broad applicability and reach. Blaze...

