NimbleEdge Open-Sources the Future of AI, Launches World's First On-Device Agentic AI Platform
With the launch of DeliteAI, NimbleEdge is putting a full-stack, production-ready on-device AI platform into the hands of developers and ML engineers. The open source stack enables fast, cross-platform deployment of transformer models, LLMs, and multimodal AI, all without relying on cloud infrastructure or high-end GPUs.
This comprehensive launch includes three major components: • A production-ready SDK with an optimized inference stack and the industry's first on-device Python runtime for orchestrating agentic workflows • A dedicated Agent Marketplace where developers can discover and integrate pre-built AI agents in their mobile applications • The NimbleEdge Assistant, the world's first fully on-device conversational AI assistant with built-in productivity capabilities 'We believe AI shouldn't sit miles away from us in data centres ,' said Varun Khare, Co-Founder and CEO of NimbleEdge. 'With this launch, any mobile application can scale AI to billions of users while improving data safety and user privacy. They own the entire stack, from the models they use to the way intelligence shows up in their products bringing a unique AI enabled experience to their users.' Designed to address the limitations of cloud-based AI, which requires constant connectivity, increases latency, creates privacy risks, and incurs unsustainable operational costs, NimbleEdge also fills critical gaps in the existing on-device AI ecosystem. Until now, developers lacked unified tooling, standardized runtimes, and a robust marketplace of ready-to-integrate agents to build sophisticated AI-native experiences on smartphones.
NimbleEdge's platform empowers any company or developer to bring their own models, including Llama, Gemma, or Qwen, and run all inference directly on the user's device, abstracting away the complexities of diverse mobile hardware and managing runtimes like ONNX, LiteRT, or ExecuTorch. This architecture ensures that no personal data ever leaves local hardware, allowing organizations to fine-tune and deploy large language models and agents entirely offline. Unlike proprietary assistants tied to a single ecosystem, NimbleEdge makes it possible to create completely customized workflows and branded assistants all powered by an open-source, on-device ecosystem without external dependencies.
Neeraj Poddar, Co-Founder and CTO at NimbleEdge, added, 'For the first time, developers can bring state-of-the-art AI models to consumer devices, orchestrate them with Python, and deploy truly private AI agents at scale. This is the missing infrastructure layer and developer tooling we wished existed when we were building distributed systems at global scale, and now it's open for everyone.' NimbleEdge has already demonstrated this capability at scale, powering AI & ML experiences across more than 30 million devices in production deployments for gaming and e-commerce apps, with AI infrastructure partners such as PyTorch and ONNX. The platform's on-device Python runtime allows developers to build dynamic, real-time applications with familiar tools while maintaining full control over data flows, and the Agent Marketplace provides a growing library of plug-and-play agents for tasks like summarization, recommendations, and speech processing.
Aakrit Vaish, early investor in NimbleEdge and a member of the India AI Mission, commented, 'NimbleEdge reflects the innovative work happening in India's AI ecosystem. By bringing agentic AI capabilities on-device, they're addressing important challenges around privacy and latency and creating open source infra that can reach users at scale in India and globally.' India's AI Mission aims to build sovereign, privacy-preserving AI infrastructure that can scale to 1.4 billion people. NimbleEdge's open, on-device platform aligns with this vision by enabling AI to run natively on India's vast base of smartphones, reducing reliance on scarce data center compute and ensuring personal data stays on users' devices. By complementing public initiatives like UPI, ONDC, and the DPDP Act, NimbleEdge can help accelerate India's leadership in trusted, accessible AI.
The NimbleEdge Platform, Agent Marketplace, and Assistant are available today. Developers can explore the source code on GitHub and join the DeliteAI Discord community to connect and collaborate. Enterprises can engage NimbleEdge for deployment support and advanced features. NimbleEdge is committed to redefining the future of AI with an open, privacy-first infrastructure that democratizes access to advanced intelligence on billions of devices worldwide.
Website: https://www.nimbleedge.com/ (Disclaimer: The above press release comes to you under an arrangement with NRDPL and PTI takes no editorial responsibility for the same.).
This is an auto-published feed from PTI with no editorial input from The Wire.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Economic Times
20 minutes ago
- Economic Times
Giant Wyoming data center to guzzle 5x more power than residents, but the user remains secret
Wyoming AI data center power consumption: A huge AI data center is planned in Cheyenne, Wyoming. It needs massive power, more than all Wyoming homes use. The center will have its own power source. The user is secret, but some speculate it is OpenAI. Experts think it will house advanced AI hardware. The project could affect Wyoming's energy landscape and utility prices. Tired of too many ads? Remove Ads Massive AI Data Center's Power Demand to Exceed Wyoming's Residential Power Use Identity of Data Center's End User Remains a Mystery Tired of too many ads? Remove Ads Is OpenAI Behind the Project? Experts Predict High-Performance AI Hardware Inside Facility Potential Ripple Effects on Utility Prices and Energy Exports Tired of too many ads? Remove Ads FAQs A massive new AI data center planned for Cheyenne, Wyoming, is making waves, not just because of its massive power requirements, but also due to the mystery surrounding who will actually use it, as per a proposed facility, a joint project between energy company Tallgrass and AI data center developer Crusoe, is expected to start with a power draw of 1.8 gigawatts, with the potential to scale up to 10 gigawatts, as reported by TechRadar. To put that in perspective, that's over five times the electricity used by every household in Wyoming combined, according to the to the huge energy demand, this new data center won't be tapping into the public power grid, as per TechRadar. Instead, the developers will reportedly build a dedicated energy supply, likely a mix of natural gas and renewable sources, to run the facility independently, according to the report. That might ease concerns about local blackouts, but it hasn't quieted all READ: Disney+ cancels Goosebumps series after two seasons: What's next for the horror anthology? Adding intrigue to the project is the complete secrecy about the end user, as per the TechRadar report. While neither Tallgrass nor Crusoe has confirmed who the data center is being built for, speculation is are many speculations that point to OpenAI as the likely tenant as the AI giant has a history with Crusoe, recently teaming up for facility in Texas, described as the 'largest data center' in the world, as reported by TechRadar. That project reportedly draws about a gigawatt of energy, which is part of OpenAI's broader 'Stargate' initiative, according to the has also been public about plans to build several gigawatts of data center capacity in the near future, but has not announced any presence in Wyoming so far, as per the TechRadar report. Crusoe has declined to comment to TechRadar on whether this new Cheyenne project is tied to Stargate or OpenAI at all, fueling more curiosity and speculation, as per the READ: Apple rumored to launch 12.9-inch MacBook this year at a price that will shock you - here are the details What is known is that this facility will likely house some of the most advanced AI hardware as industry experts expect it to house the fastest CPUs available, possibly in dense, rack-mounted workstation configurations optimized for deep learning and model training, as reported by to the report, these systems are power-hungry due to the design because each server node is capable of handling massive workloads that demand sustained cooling and uninterrupted READ: Texas AI centers guzzle 463 million gallons, now residents are asked to cut back on showers While Wyoming state officials have welcomed the project, touting its potential to support local industries, especially natural gas, but others are raising red flags, as per the TechRadar if the facility powers itself, experts warn it could impact the broader energy landscape, as per the report. There are concerns that residents of Wyoming and its locality could face higher utility costs, especially if local supply chains or pricing models are indirectly affected, and even Wyoming's identity as a major energy exporter could be tested if more such facilities emerge, as per the TechRadar one knows for sure yet. The companies involved haven't named the client, but many believe it could be not directly. The facility will have its own energy supply, separate from the public grid.
Time of India
36 minutes ago
- Time of India
Giant Wyoming data center to guzzle 5x more power than residents, but the user remains secret
Wyoming AI data center power consumption: A huge AI data center is planned in Cheyenne, Wyoming. It needs massive power, more than all Wyoming homes use. The center will have its own power source. The user is secret, but some speculate it is OpenAI. Experts think it will house advanced AI hardware. The project could affect Wyoming's energy landscape and utility prices. Tired of too many ads? Remove Ads Massive AI Data Center's Power Demand to Exceed Wyoming's Residential Power Use Identity of Data Center's End User Remains a Mystery Tired of too many ads? Remove Ads Is OpenAI Behind the Project? Experts Predict High-Performance AI Hardware Inside Facility Potential Ripple Effects on Utility Prices and Energy Exports Tired of too many ads? Remove Ads FAQs A massive new AI data center planned for Cheyenne, Wyoming, is making waves, not just because of its massive power requirements, but also due to the mystery surrounding who will actually use it, as per a proposed facility, a joint project between energy company Tallgrass and AI data center developer Crusoe, is expected to start with a power draw of 1.8 gigawatts, with the potential to scale up to 10 gigawatts, as reported by TechRadar. To put that in perspective, that's over five times the electricity used by every household in Wyoming combined, according to the to the huge energy demand, this new data center won't be tapping into the public power grid, as per TechRadar. Instead, the developers will reportedly build a dedicated energy supply, likely a mix of natural gas and renewable sources, to run the facility independently, according to the report. That might ease concerns about local blackouts, but it hasn't quieted all READ: Disney+ cancels Goosebumps series after two seasons: What's next for the horror anthology? Adding intrigue to the project is the complete secrecy about the end user, as per the TechRadar report. While neither Tallgrass nor Crusoe has confirmed who the data center is being built for, speculation is are many speculations that point to OpenAI as the likely tenant as the AI giant has a history with Crusoe, recently teaming up for facility in Texas, described as the 'largest data center' in the world, as reported by TechRadar. That project reportedly draws about a gigawatt of energy, which is part of OpenAI's broader 'Stargate' initiative, according to the has also been public about plans to build several gigawatts of data center capacity in the near future, but has not announced any presence in Wyoming so far, as per the TechRadar report. Crusoe has declined to comment to TechRadar on whether this new Cheyenne project is tied to Stargate or OpenAI at all, fueling more curiosity and speculation, as per the READ: Apple rumored to launch 12.9-inch MacBook this year at a price that will shock you - here are the details What is known is that this facility will likely house some of the most advanced AI hardware as industry experts expect it to house the fastest CPUs available, possibly in dense, rack-mounted workstation configurations optimized for deep learning and model training, as reported by to the report, these systems are power-hungry due to the design because each server node is capable of handling massive workloads that demand sustained cooling and uninterrupted READ: Texas AI centers guzzle 463 million gallons, now residents are asked to cut back on showers While Wyoming state officials have welcomed the project, touting its potential to support local industries, especially natural gas, but others are raising red flags, as per the TechRadar if the facility powers itself, experts warn it could impact the broader energy landscape, as per the report. There are concerns that residents of Wyoming and its locality could face higher utility costs, especially if local supply chains or pricing models are indirectly affected, and even Wyoming's identity as a major energy exporter could be tested if more such facilities emerge, as per the TechRadar one knows for sure yet. The companies involved haven't named the client, but many believe it could be not directly. The facility will have its own energy supply, separate from the public grid.
Indian Express
2 hours ago
- Indian Express
How Artificial Intelligence is helping badminton players improve their games
There is an epidemic of poorly executed forehand air tosses in badminton, amongst the hundreds of player videos collated in Bengaluru. A bug was blighting the feature. An AI-powered review system had picked up the trend pointing to the weakness of that high altitude shot amongst upcoming shuttlers, mostly beginners. It has zeroed in on the elbow not being properly stretched and the balance being slightly off, suggesting foot placement and arm control drills to a specific shuttler. All that the shuttler had done was upload a video of a training session/match, and the VisistAI tech from a Bengaluru startup, reeled off a basic analysis of the toss – one of the tactical weapons deployed in matches that demands consistency and accuracy on the time of contact of shuttle with racquet. The tech startup is attempting to fill up a gap for grassroots badminton players by offering them access to biomechanic tools that analyse their technique and suggest corrections. 'One insight that was massive from the VisistAI review was that several players had a weak forehand toss, and were scoring less from that shot. It was an instruction to coaches to work with strength & conditioning coaches. Our tools provide elite insights at various stages of play by assessing skill, strokes and agility,' says Suresh Chintada, co-founder and CEO of VisistAI. Players record their games on a smartphone mounted on a tripod, and send it off for AI model to threadbare their techniques. Typically, a rudimentary report throws up reviews on assorted skills like grip, the forehand keep shot, backhand serves besides the toss. 'Besides agility, we can track specific postures, movements on court, and shuttle flights,' explains Seshadri Raghavan, co-founder and chief tech officer. Almost 70+ Bengaluru academies are availing of this service, ensuring a single coach can actually keep track of several games. 'We have assessment reports and smart journals that we send to the coaches, and AI acts like an assistant coach,' Suresh says. Searchable bookmarked footage with dashboards showing historical trends and improvements tracked over several weeks, is available even as the AI assessment uses Rubric templates (predefined parameters in boxes), and reports are available in local languages. An almost imperceptible smashing technique tweak once came up when the AI model noted, 'Preparation of elbow and arm is too low before smashing. In under pressure situations it is difficult to get good angles on the smash.' A simple grip – though all-important- is assessed for consistent relaxed hold, finger placement, follow through of wrist, while pointing out that tiny foot shifts tend to disturb execution. 'Improve explosiveness to reach the shuttle sooner. Work on smoother timing on serves and reinforce elbow alignment in air toss,' It summarises. Grip relaxing exercises on the odd occasion it gets too tight and impedes a stroke, are advised. For a particular shuttler, the AI model pulled out Frame No 376 for ideal shot placement on a forehand keep shot, while shuttle placement is a weakness. On the backhand serve, the player was advised steadying the stance even if it's only in a couple of instances. 'For an experienced coach, the AI model is an augmented, unblinking view with a couple of dimensions added to look at movement, posture,' says Suresh who adds that coaches have returned to tell him their level of coaching and numbers of competitive players has gone up with access to this information which is now objectively and democratically available – though at a price. The duo always fancied sports – tennis and even running technique of marathons, though VisistAI is available for similar analyses in cricket and basketball. But all AI developments need to be taken with a pinch of salt, and the duo insist no AI can replace a coach. 'AI will never replace a coach. It's not in a position to understand mental aspects like a 'bad day' which can have a massive impact on the game. But now anybody can become a coach and access this,' Suresh adds. Access to elite coaching is 0.1 percent in India and 25 percent in the USA, and AI models with workflow charts can help bridge these gaps. The startup hopes to add human expert coaches to the panel to go a step further in making advice available. But the ambitions ultimately extend to predictive analysis, and the final frontier is stopping injuries from happening. 'Injury prevention is the goal. Right now tech is used in biomechanics and rehab level. But if we can predict injury before it happens and reduce chronic stress ..' says Seshadri.
