Krutrim Starts Hosting Meta’s Llama 4 Models On Its Cloud In India

SUMMARY

Krutrim said that Llama 4 Scout and Llama 4 Maverick will be available for developers to build and deploy applications in the price range of INR 7 to INR 17 per Mn tokens

With this, the unicorn claims to have become the first Indian AI company to deploy Meta’s Llama 4 models on Indian servers

The startup also reiterated that the second version of Krutrim Assistant is well on track to be rolled out later this month and it will feature the “DeepSearch” tool

Bhavish Aggarwal-led artificial intelligence (AI) unicorn Krutrim has said that it has started hosting Meta’s Llama 4 models on its cloud platform.

In a statement, Krutrim said that Llama 4 Scout and Llama 4 Maverick will be available for developers to test, build, and deploy applications in the price range of INR 7 to INR 17 per Mn tokens.

While Llama 4 Scout Model is a 17 Bn active parameter model with 16 experts and 10 Mn token context, Llama 4 Maverick boasts 17 Bn active parameters with 128 experts and 1 Mn token context window.

For context, tokens are a fundamental unit of text that LLMs generate.

With this, the unicorn claims to have become the first Indian AI company to deploy Meta’s Llama 4 models on Indian servers.

“Excited to share that @Krutrim is among the world’s first to host Meta’s Llama 4 models running entirely on it’s India-hosted cloud. Powering our developers with world-class AI, at industry-disrupting prices, with complete data sovereignty,” said Aggarwal in a post on X.

In a statement, the startup reiterated that the second version of Krutrim Assistant is well on track to be rolled out later this month. It also said that the Assistant, V2, will feature “DeepSearch”, a tool designed to make data searches more precise and efficient.

The AI startup also reiterated plans to scale up its data centre capacity to 1 GW by 2028.

The development comes a few months after the Bhavish Aggarwal-led AI unicorn, in January 2025, began hosting open source AI models of Chinese GenAI company DeepSeek on its cloud platform. A month later in February, it also deployed DeepSeek’s new R1 671B model on Nvidia’s H100 graphics processing units in India.

Founded in 2023, Krutrim offers GPU-as-a-service, model-as-a-service, along with other multiple no-code platforms. It became a unicorn in January 2024 after raising $50 Mn in a round led by Z47 (erstwhile Matrix Partners India).

The startup has also submitted a proposal to the government to build indigenous AI foundational models under the INR 10,037 Cr IndiaAI Mission. Earlier this week, IT minister Ashwini Vaishnaw said that the evaluation process of the proposals is in the “final leg”, adding that the first few of the selected startups will be offered funding by the Centre.