Building Trusted AI at Scale Cities Startups & Digital Sovereignty – Keynote Vivek Raghavan Sarvam AI

20 Feb 2026 13:00h - 14:00h

Session at a glance

Summary

Vivek Raghavan, co-founder of Sarvam AI, delivered a keynote presentation emphasizing India’s capability to build sovereign artificial intelligence technology that serves the country’s diverse linguistic and cultural needs. Raghavan argued that developing indigenous AI capabilities is not optional but essential for India to avoid becoming a “digital colony” dependent on foreign technology, drawing parallels to India’s successful development of the Aadhaar digital identity system and India Stack. He highlighted India’s unique advantages in AI development, including its linguistic diversity with 22 official languages, its position as a large economy with significant demand for AI services, and its cost-conscious approach that drives efficient solutions.

Sarvam has developed a full-stack sovereign AI platform consisting of three components: models, applications, and infrastructure. The company has created several state-of-the-art models built entirely from scratch, including the SARAS speech recognition model that outperforms global competitors in Indian languages, advanced text-to-speech and dubbing capabilities, and vision models for document digitization. Their language models include a 30-billion parameter conversational model and a 105-billion parameter model, which represents the largest LLM trained from scratch in India. These models demonstrate performance superior to comparable global models while being significantly smaller and more efficient.

The company has successfully deployed these models in real-world applications, processing over one million minutes of voice conversations daily across 11 Indian languages and working with enterprises, government agencies, and NGOs. Raghavan emphasized that this achievement was accomplished by a team of just 15 young developers, demonstrating India’s potential to lead in AI development with proper support and resources.

Keypoints

Major Discussion Points:

– AI Sovereignty and National Independence: The critical need for India to develop its own AI capabilities rather than depending on foreign technologies, drawing parallels to how India built its own digital identity system (Aadhaar) and payment infrastructure (UPI) to avoid becoming a “digital colony”

– India’s Unique Advantages in AI Development: Leveraging India’s linguistic diversity (22 official languages, dialects changing every 50 kilometers), large economy with high demand for AI services, cost-conscious market requiring efficient solutions, and abundant developer talent

– Sarvam’s Sovereign AI Models and Technical Achievements: Demonstration of world-class AI models built from scratch in India, including speech recognition (SARAS), text-to-speech, vision models, and large language models (up to 105 billion parameters) that outperform global competitors, especially in Indian languages

– Practical Applications and Real-World Impact: Converting research into usable applications including real-time voice conversations in 11 Indian languages (processing over 1 million minutes daily), content digitization, translation services, and enterprise solutions

– Infrastructure and Scalability Vision: Building comprehensive AI infrastructure that can operate at “India scale” and “India cost,” including edge computing capabilities, innovative form factors like AI glasses, and partnerships with enterprises and government agencies

Overall Purpose:

This keynote presentation aims to demonstrate that India has the capability to build world-class AI technology independently, showcasing Sarvam’s achievements as proof that the country can compete globally while serving its unique linguistic and cultural needs.

Overall Tone:

The discussion maintains a consistently inspirational and patriotic tone throughout, with Raghavan positioning himself as a champion of Indian technological capability. The tone is confident and optimistic, emphasizing empowerment (“India can”) while also carrying an urgent undertone about the necessity of AI sovereignty for national security and independence. The speaker balances technical demonstrations with motivational messaging, maintaining enthusiasm even when technical demonstrations encounter minor glitches.

Speakers

– Announcer: Role/Title: Event announcer; Area of expertise: Not mentioned

– Vivek Raghavan: Role/Title: Co-founder of Sarvam; Area of expertise: AI development, digital identity systems (worked on building Aadhaar), sovereign AI technology, Indian language AI models

Additional speakers:

None identified beyond the speakers names list.

Full session report

Vivek Raghavan, co-founder of Sarvam AI, delivered an inspiring keynote presentation centered on his core message that “India can” build world-class sovereign artificial intelligence capabilities. His presentation demonstrated how Sarvam AI has developed a comprehensive full-stack AI platform that proves Indian teams can achieve cutting-edge results while serving India’s unique linguistic and cultural context.

The Sovereignty Imperative

Raghavan framed AI sovereignty as essential for India’s technological independence, warning that without indigenous AI capabilities, India risks becoming dependent on foreign technology for core functions. Drawing from his experience working on building Aadhaar over the past 15 years, he argued that while the world focuses on having the largest models or fastest chips, sovereignty ultimately determines a nation’s long-term position in the technological landscape.

He positioned this within India’s successful track record of building proprietary, open-source public infrastructure, citing the transformation to India Stack as evidence of the country’s capability to achieve technological independence. Raghavan emphasized that AI’s pervasive impact on every aspect of human life makes indigenous development absolutely essential.

India’s Strategic Advantages

Rather than viewing India’s complexity as a challenge, Raghavan presented it as the country’s greatest competitive advantage. He highlighted India’s remarkable linguistic diversity—22 official languages with variations every 50 kilometers—as creating unique datasets and use cases that global models cannot adequately address. This diversity, he argued, must be captured to truly understand “the voice of the people.”

Raghavan identified India’s additional strategic advantages: its position as one of the world’s largest economies with substantial AI demand, cost-conscious market dynamics that drive efficient innovation, and its vast developer community. He drew parallels to UPI’s success, noting that Indians could feel domestic technology was superior to global alternatives—a psychological shift that creates the foundation for AI leadership.

Sarvam’s Full-Stack Platform

Raghavan explained that Sarvam has built a comprehensive three-part platform: models, applications, and infrastructure. Crucially, all models are developed entirely from scratch without dependency on external systems, while maintaining world-class performance standards.

Technical Achievements and Demonstrations

Despite experiencing technical difficulties with audio and video demonstrations during his presentation, Raghavan detailed Sarvam’s impressive technical accomplishments:

SARAS Speech Recognition: The model demonstrates superior performance in Indian languages compared to global competitors while being significantly smaller. Its native understanding of Indian linguistic nuances enables it to outperform larger international systems in blind tests.

Text-to-Speech and Dubbing: Sarvam’s voices have been recognized as the most preferred in Indian languages when compared to established competitors like Eleven Labs and Cartesia. The platform offers mixed language dubbing capabilities and studio-quality content solutions.

Vision Models: Their 3 billion parameter state space model achieves world-class performance while being orders of magnitude smaller than competing systems. The model excels not only in Indian languages but matches or exceeds performance in English.

Language Models: Sarvam has developed both a smaller, efficient model and a flagship large language model. The smaller model is “extremely small” and “can run on a single GPU” with “32K context length and is trained on 16 trillion tokens.”

The crown jewel is their 105 billion parameter large language model—the largest LLM trained from scratch in India. This model performs on par with leading alternatives like “GBT OSS 120 billion” and “Gemini Flash” while maintaining superior capabilities in Indian languages. Remarkably, Raghavan noted their performance metrics are “superior to what DeepSeek R1 had last year,” despite DeepSeek R1 being “670 billion parameters.”

Real-World Scale and Impact

Raghavan emphasized that these achievements extend beyond research to practical applications serving real users. The platform processes over one million minutes of voice conversations daily across 11 Indian languages, working with “many leading enterprises,” government agencies, and NGOs.

The company’s recent collaboration with 20 NGOs processed “a crore minutes of calling” “in the last month,” demonstrating the platform’s capacity to operate at national scale. Beyond voice applications, Sarvam has developed comprehensive solutions including book digitization, translation services, and video dubbing capabilities.

Infrastructure and Accessibility

Sarvam’s infrastructure strategy addresses practical deployment challenges across India’s diverse geography and economic spectrum. The company is optimizing models for edge computing and mobile deployment, including innovative form factors like AI-enabled glasses that can process Indian voices across different contexts.

Raghavan credited the India AI Mission grant as essential for their LLM development, specifically noting the GPU grant that enabled their large model training.

Youth Empowerment and Team Achievement

Perhaps most remarkably, Raghavan emphasized that Sarvam’s world-class models were developed by a team of just “15 young people,” positioning himself as merely “the spokesperson” for their collective achievement. He framed this as part of “the game of the youth of India,” suggesting that with proper support, India’s talent pool could achieve transformational results.

Conclusion: A Beginning, Not an End

Raghavan concluded by stating that “the story of Sarvam is just a beginning,” suggesting enormous potential for India’s AI future. His presentation successfully demonstrated that Indian teams can build state-of-the-art AI systems that serve both national needs and achieve global competitiveness.

The keynote provided concrete evidence that India’s AI ambitions are not merely aspirational but achievable, with Sarvam’s technical accomplishments serving as proof that indigenous innovation can turn the country’s linguistic and cultural complexity into competitive advantages in the global AI landscape.

Session transcript

Announcer

I move on to our next keynote speaker, who is Mr. Vivek Raghavan, the co -founder of Sarvam, a company building AI that speaks India’s languages and understands India’s context. In a world dominated by models trained on English language data, their work is a powerful demonstration that sovereign AI capability is not just a luxury, it is a necessity. So, ladies and gentlemen, please welcome Mr. Vivek Raghavan, co -founder of Sarvam.

Vivek Raghavan

I come here to say that India can. And I think that’s the message I want to say. India can. And India can train state -of -the -art models, bring AI to a billion Indians, and do it all. And that’s really the message of why we started Sarvam. I want to talk about the long arc. You know, when you look at, you know, today the world is moving so fast. Everybody talks about where is the largest model or who has the fastest chip. These are all transitory technical advantages. In the long run, it’s sovereignty that matters. And unless we build these things ourselves, it’s something that, you know, will be left behind in the race. In the past 15 years of my life, I worked on building Aadhaar.

Which is India’s… digital identity program. Prior to that, many of these technologies, many of these technologies were proprietary technologies. And we built this kind of self -created technology that is open source and a public infrastructure that is available to all of us. And that led to the creation of the India Stack. So when you look at it over the course of long periods of time, sovereignty will always trump technical leads that are short term. We have a mandate to build. It is not an option whether we want to build in these technologies. AI is a technology that has impact on every single aspect of human life. And it’s a core technology that a country like India must have.

And it’s a core technology that a country like India must understand. from the foundational level. Otherwise, we will become a digital colony which is dependent on other countries for this core, core technology. That’s something that is, it’s not an option. It is something that we must do. And we have unique advantages. And our unique advantage is actually our diversity. We have so many languages. We have 22 official languages. And in fact, you know, the way people speak in our country changes every 50 kilometers. And that diversity must be captured if we have to understand the voice of the people. And therefore, if we build AI from India, it must acknowledge that diversity and do this.

The other thing, of course, is we are a huge economy. There is demand. and if AI is there to help the citizen to do everything, this is we can be one of the largest consumers of AI in the world. And that demand is there and then we have to build. We know that we are a cost -conscious country, right? Everything needs to be at the lowest cost. So we need to build efficient AI that actually can be delivered at scale for the people so that the last person in the country can actually have a better experience, right? Today, if you look at UPI, one of the great success stories of the past decade, and if you look, it is for the first time we feel that in India, things can be better than everywhere else in the world.

But AI done the right way can make sure that every service to citizens actually is the best and the cheapest and actually done in the best possible way for the country and that’s the promise of AI and that’s why we said we need to do this in India. and I think it’s not about I mean we have companies which have globally when you look at AI companies they are massive companies but in the end we have to show a new model where we can actually build AI which helps everyone and then we can win for the people and our model can be adopted in the world and that is my belief of where we need to go on this thing So Sarvam has been building India’s full stack sovereign AI platform and we work with developers fundamentally India is a country of developers we have more developers and we work with enterprises and we work with governments and we have a full stack platform which I’ll talk about In fact the full stack platform consists of three different things one is models we need models that are built in India And that is the key thing and that’s what we’ll focus on, sovereignty and models.

And then we are focusing on applications. Applications is AI for everyday tasks, for making things better for people. And finally, we’ll talk about infrastructure and infrastructure at India scale. The first thing I want to talk about are sovereign models. Rule number one is they are built from scratch. They are not dependent on any other model that is there in the world. They are built from scratch. There is no data dependency on anyone else. But at the same time, the focus is these are world -class, state -of -the -art models that we have. So I’m going to talk a little bit about some of our models. The first model is actually the SARAS model, which is actually the speech model which helps recognize speech in Indian languages.

This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. This is native. In fact, if you see, this is actually best in class in Indian languages compared to any other global model in terms of that’s something to say that in the, these are extremely small models, but these are models which have been trained with significant amounts of Indian diverse data, which will actually lead to better performance on Indie voices.

So let me just take a play, an iconic moment in India, diarized using our model. Oh, sorry. Okay, I don’t think I can make this happen. Okay. Is audio, it’s not playing. let’s maybe we’ll come back to this and here we have a majestic liftoff of lbm3 m4 rocket carrying india’s prestigious chandrayaan 3 spacecraft this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket this is the rocket of lbm3 m4 rocket We want to create models which are naturally expressive Indian voices, and they have low latency streaming and actually production grade quality.

So, in fact, our models, our speech to text models are considered in blind tests are actually the most preferred voices in Indian languages. And this is something compared to all the global competitors, such as Levin Labs and Cartesia, etc. And we have actually the most preferred text to speech model in the country. We’ll also talk, we also have a dubbing capability, which actually preserves the speaker modality and has precise control over duration and supports mixed language things. So I will show, in fact, in this model as well. We see compared to any other model in the world, we are the most preferred as far as dubbing is concerned. I will show a small snippet of what happens here.

We have the remaining 13 bits in which the 12th bit is called the small a bit. Then the remaining 6 bits are compute bits. These 3 are the destination bits and these are the jump bits. Then we have the remaining 13 bits in which the 12th bit is called the small a bit. Then the remaining 6 bits are compute bits. These 3 are the destination bits and these are the jump bits. Then we have the remaining 13 bits in which the 12th bit is called the small a bit. Then the remaining 6 bits are compute bits. These 3 are the destination bits and these are the jump bits. So we’ve also built these vision models.

And these vision models are actually very good for document digitization. They’re very good at language layout understanding, visual grounding, and in fact, finding intelligence by visual components. And finally, reading order predictions. In fact, the vision model that we built is only a 3 billion parameter state space model which beats all other models in the world, not just in Indian languages, but in English as well. So therefore, it shows, and many of these models are many orders of magnitude bigger than our models, and still we are able to get world -class performance from them. Of course, in Indian languages, we are far and away ahead of the models that are there from the global comparison. Now we come talk about some of the LLMs.

We have actually, India has started the training of LLMs from scratch. And this was done through a grant from the India AI Mission that actually without which it would not have been possible for us to train these kinds of models through a GPU grant. In fact, it is a context, it’s a model which is an extremely small model which can run on a single GPU. And it has a 32K context length and is trained on 16 trillion tokens. And it’s extremely efficient thinking which actually gives better answers with lower. And the focus of this model is actually real time conversational applications that people will be able to, it will be able to generate conversations in all the Indian languages in a real time system.

And some of the benchmarks show compared to models, global models of similar size. such as QN30B or GPT -OSS. It is far superior in the ability in terms of various parameters, such as fluency, language and script, and usefulness and conciseness. So the important thing is that this is able to, at the size, it is again a global best. And then we’ll come finally to our largest model, which is the 105 billion parameter model. This is the largest LLM that has trained from scratch in the country. And it is basically better on par with most open source and closed source models of its class. And it can handle various kinds of complex reasoning tasks, as well as web searching and things like that.

So it’s actually a fairly complex, fairly advanced model, which again works in all… all Indian languages. From a benchmark perspective, these models, again, compared to things like GBT OSS 120 billion, as well as Gemini Flash, is actually superior in terms of the kinds of outputs it can generate. So therefore, this is really an example. While this is quite a small model, it is really, just to give you an idea, last year we had DeepSeek R1, which actually was launched, and that was 670 billion parameters. The numbers that we are getting are actually superior to what DeepSeek R1 had last year. Of course, the state of the art has also improved. But the goal is we can show that India can build these things.

And I want this. The most important thing that I want people to understand is… just because, and I think that the, you know, I would love that not just us, but many other people come and show that we can actually build world -class models from India, because that is the fact. And these models was built with a team of just 15 young people. And really it’s the game of the youth of India that have actually made this model what it is. I’m just the spokesperson. These young kids have actually made something like this happen. And if these kids can do it, we have so much talent in the country. And I think that’s, I’m very positive about given the right kind of support in the way that we have been given, that much bigger things can happen.

Moving beyond our models, we actually want these models to become useful. And so therefore we build applications. And I’ll talk very briefly. about the kinds of applications that we build. So in fact, we have an ability to actually converse, and this is our real -time voice conversation that happens. We do more than a million minutes of voice conversation in 11 Indian languages every day using Servum. So these models are actually being used to build things. So these models which have been trained fully in our control are actually now being used for conversations across enterprises and government use cases. In fact, in the last month, we actually took about 20 NGOs, actually did a crore minutes of calling within a month to really understand what is the real -time voice conversation and what people actually are saying at the ground.

Okay. So we actually have ways by which we can make this available for work tasks and enterprise tasks. And we also have the ability to do this for content, which is digitizing books, translating books, dubbing videos. These are all studio products that we have. And I think finally, I want to end with infrastructure. We actually are doing many interesting things. We are making our models smaller to work on the edge, to work on phones. We are making, we’re actually, many of you may have heard that we have also launched some of these glasses so that we are able to have these models run on different form factors and capture the intelligence, capture the voice of India at every point in there.

And finally, for these things to work, you need to have compute at a large scale and the ability to actually very efficiently deliver all these models to India at India scale at India cost. Finally, we are, of course, trusted by many leading enterprises, work with many government agencies, and I think the story of the Servum is just a begining

Announcer

Speech speed

133 words per minute

Speech length

73 words

Speech time

32 seconds

Sovereign AI capability is a necessity, not a luxury

Explanation

The announcer frames sovereign AI as an essential requirement for nations rather than an optional advantage. This positions AI independence as a strategic imperative in a landscape dominated by English‑language models.

Evidence

“In a world dominated by models trained on English language data, their work is a powerful demonstration that sovereign AI capability is not just a luxury, it is a necessity.” [3]

Major discussion point

AI Sovereignty as Strategic Necessity

Topics

Artificial intelligence

Vivek Raghavan

Speech speed

140 words per minute

Speech length

Infrastructure and Future Outlook

Topics

Artificial intelligence | Capacity development | Financial mechanisms

Agreements

Agreement points

AI sovereignty is essential for national independence

Speakers

– Vivek Raghavan
– Announcer

Arguments

AI sovereignty is essential for India to avoid becoming a digital colony dependent on other countries for core technology

Sarvam demonstrates that sovereign AI capability is not just a luxury but a necessity

Summary

Both speakers agree that developing indigenous AI capabilities is not optional but essential for maintaining national technological independence and avoiding dependency on foreign technology

Topics

Artificial intelligence | The enabling environment for digital development

Similar viewpoints

Both speakers emphasize that sovereign AI development is a necessity rather than a luxury, with the announcer introducing this concept and Raghavan providing detailed justification through his experience and company’s work

Speakers

– Vivek Raghavan
– Announcer

Arguments

AI sovereignty is essential for India to avoid becoming a digital colony dependent on other countries for core technology

Sarvam demonstrates that sovereign AI capability is not just a luxury but a necessity

Topics

Artificial intelligence | The enabling environment for digital development

Unexpected consensus

Complete alignment on AI sovereignty necessity

Speakers

– Vivek Raghavan
– Announcer

Arguments

AI sovereignty is essential for India to avoid becoming a digital colony dependent on other countries for core technology

Sarvam demonstrates that sovereign AI capability is not just a luxury but a necessity

Explanation

The unexpected consensus lies in the fact that both the announcer (typically neutral) and the speaker are in complete alignment on the critical importance of AI sovereignty, with no dissenting voices or alternative perspectives presented in the discussion

Topics

Artificial intelligence | The enabling environment for digital development

Overall assessment

Summary

There is complete consensus between the two speakers on the fundamental importance of AI sovereignty for national independence, with both emphasizing it as a necessity rather than an option

Consensus level

The level of consensus is extremely high, with no disagreements or alternative viewpoints presented. This strong alignment reinforces the message that indigenous AI development is critical for avoiding technological dependency and maintaining national sovereignty in the digital age

Differences

Different viewpoints

Unexpected differences

Overall assessment

Summary

No disagreements identified in the transcript

Disagreement level

This transcript contains only one substantive speaker (Vivek Raghavan) presenting his company’s work and vision for AI sovereignty in India. The announcer merely provides an introduction that aligns with Raghavan’s message. There are no opposing viewpoints, counterarguments, or alternative approaches presented. This represents a keynote presentation format rather than a debate or discussion with multiple perspectives.

Partial agreements

Similar viewpoints

Both speakers emphasize that sovereign AI development is a necessity rather than a luxury, with the announcer introducing this concept and Raghavan providing detailed justification through his experience and company’s work

Speakers

– Vivek Raghavan
– Announcer

Arguments

AI sovereignty is essential for India to avoid becoming a digital colony dependent on other countries for core technology

Sarvam demonstrates that sovereign AI capability is not just a luxury but a necessity

Topics

Artificial intelligence | The enabling environment for digital development

Takeaways

Key takeaways

India must develop sovereign AI capabilities to avoid becoming a digital colony dependent on other countries for core technology

Long-term sovereignty in AI is more important than short-term technical advantages in the global competition

India’s linguistic diversity (22 official languages with regional variations) provides unique advantages for AI development

Sarvam has successfully demonstrated that India can build world-class AI models from scratch, including speech recognition, vision models, and large language models

Small teams can achieve significant results – Sarvam’s models were built by just 15 young people

Practical applications are already at scale with over 1 million minutes of voice conversations processed daily in 11 Indian languages

India’s cost-conscious approach and large developer community position it well for efficient AI development and deployment

AI built in India must acknowledge and capture the country’s diversity to truly understand the voice of the people

Resolutions and action items

Continue building India’s full-stack sovereign AI platform across models, applications, and infrastructure

Scale real-time voice conversation capabilities across enterprises and government use cases

Develop edge computing capabilities and alternative form factors like glasses for broader accessibility

Make AI models smaller and more efficient to work on phones and edge devices

Build compute infrastructure at large scale to deliver models efficiently across India

Unresolved issues

How to encourage more companies and teams beyond Sarvam to build world-class AI models in India

Specific strategies for scaling AI development talent and resources across the country

Long-term funding mechanisms beyond the India AI Mission grant for continued model development

Detailed roadmap for achieving cost-effective AI delivery to reach ‘the last person in the country’

Specific metrics and benchmarks for measuring success of sovereign AI initiatives

Suggested compromises

Impact

This comment transforms the discussion from being about one company’s achievements to being about India’s broader potential. It suggests scalability – if 15 people can build state-of-the-art models, imagine what thousands of talented Indians could accomplish. This shifts focus from Sarvam’s specific success to India’s systemic capability.

Overall assessment

Explanation

While mentioning work on making models smaller for phones and edge devices, the specific technical challenges and solutions for widespread deployment need further exploration

Disclaimer: This is not an official session record. DiploAI generates these resources from audiovisual recordings, and they are presented as-is, including potential errors. Due to logistical challenges, such as discrepancies in audio/video or transcripts, names may be misspelled. We strive for accuracy to the best of our ability.

Building Trusted AI at Scale Cities Startups & Digital Sovereignty – Keynote Vivek Raghavan Sarvam AI

Summary

Keypoints

Speakers

Sovereign AI capability is a necessity, not a luxury

Explanation

Evidence

Major discussion point

Topics

Avoid digital colonisation through AI sovereignty

Explanation

Evidence

Major discussion point

Topics

Long‑term power lies in self‑built technology, not transient model size or chip speed

Explanation

Evidence

Major discussion point

Topics

AI as a core, domestically understood technology for India

Explanation

Evidence

Major discussion point

Topics

India’s linguistic diversity as AI advantage

Explanation

Evidence

Major discussion point

Topics

Cost‑conscious market drives demand for low‑cost AI

Explanation

Evidence

Major discussion point

Topics

Existing developer ecosystem and public infrastructure (Aadhaar, India Stack)

Explanation

Evidence

Major discussion point

Topics

Three‑layer stack: models, applications, infrastructure

Explanation

Evidence

Major discussion point

Topics

Models built from scratch with no external data dependency

Explanation

Evidence

Major discussion point

Topics

Efficient, edge‑ready AI for the last person

Explanation

Evidence

Major discussion point

Topics

SARAS speech model best‑in‑class Indian language recognition

Explanation

Evidence

Major discussion point

Topics

Vision model (3 B parameters) beats larger international models

Explanation

Evidence

Major discussion point

Topics

Small LLM with 32K context and 16 trillion tokens for multilingual real‑time conversation

Explanation

Evidence

Major discussion point

Topics

105 B‑parameter LLM matches/exceeds top models on reasoning and web‑search

Explanation

Evidence

Major discussion point

Topics

Real‑time voice conversation >1 million minutes daily in 11 Indian languages

Explanation

Evidence

Major discussion point

Topics

Deployments with NGOs, enterprises, government for call‑centres, digitisation, translation, dubbing

Vision model (3 B parameters) beats larger international models

Small LLM with 32K context and 16 trillion tokens for multilingual real‑time conversation

105 B‑parameter LLM matches/exceeds top models on reasoning and web‑search

Real‑time voice conversation >1 million minutes daily in 11 Indian languages