Загрузка видео...

Не удалось загрузить видео

На главную

India’s Sovereign Move - Sarvam Vision Cracks the Code across 22 Indian languages Sarvam Vision is an AI model from Sarvam AI that excels in optical character recognition (OCR) for 22 Indian languages, outperforming global models like Gemini and GPT-4o on Indic benchmarks. This achievement supports India's push for...

AugadhBhudeva's profile picture

Augadh

19,835 subscribers

20,688 просмотров • 6 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Sarvam Beats GPT-4o: India’s New AI Model Claims Top Spot in Indic Speech Sarvam AI, an Indian startup, recently launched Sarvam Audio, a speech recognition model that claims superior performance over GPT-4o Transcribe on Indic language benchmarks. This development highlights India's push for AI sovereignty in handling local linguistic nuances. Sarvam Audio supports 22 Indian languages from the Eighth Schedule, plus Indian English, with strong handling of code-mixing like Hindi-English blends. It features built-in speaker diarization for up to eight speakers and processes long-form audio such as podcasts or meetings. Trained on the IndicVoices dataset 12,000 hours from over 16,000 speakers across 208 districts it captures real-world noise and spontaneous speech. The model reportedly outperforms GPT-4o Transcribe and Gemini 3 Flash in transcription accuracy (lower Word Error Rate) on IndicVoices benchmarks for unnormalized, normalized, and code-mixed speech. Sarvam attributes this to specialization on Indian accents and patterns, unlike global models trained on Western data. Detailed public benchmarks are pending independent verification. Key Applications 🔴 Call centers and logistics for multilingual transcription. 🔴 Banking, fintech, and e-commerce for customer interactions. 🔴 Podcasts, meetings, and lectures via API for real-time or batch processing. ​ 🔴 This B2B-focused tool aligns with India's IndiaAI Mission, backed by government GPU access for sovereign LLMs. Credit : AIM Networks.
5:20

Sarvam Beats GPT-4o: India’s New AI Model Claims Top Spot in Indic Speech Sarvam AI, an Indian startup, recently launched Sarvam Audio, a speech recognition model that claims superior performance over GPT-4o Transcribe on Indic language benchmarks. This development highlights India's push for AI sovereignty in handling local linguistic nuances. Sarvam Audio supports 22 Indian languages from the Eighth Schedule, plus Indian English, with strong handling of code-mixing like Hindi-English blends. It features built-in speaker diarization for up to eight speakers and processes long-form audio such as podcasts or meetings. Trained on the IndicVoices dataset 12,000 hours from over 16,000 speakers across 208 districts it captures real-world noise and spontaneous speech. The model reportedly outperforms GPT-4o Transcribe and Gemini 3 Flash in transcription accuracy (lower Word Error Rate) on IndicVoices benchmarks for unnormalized, normalized, and code-mixed speech. Sarvam attributes this to specialization on Indian accents and patterns, unlike global models trained on Western data. Detailed public benchmarks are pending independent verification. Key Applications 🔴 Call centers and logistics for multilingual transcription. 🔴 Banking, fintech, and e-commerce for customer interactions. 🔴 Podcasts, meetings, and lectures via API for real-time or batch processing. ​ 🔴 This B2B-focused tool aligns with India's IndiaAI Mission, backed by government GPU access for sovereign LLMs. Credit : AIM Networks.

Augadh

43,429 просмотров • 6 месяцев назад

India is accelerating its sovereign AI mission with homegrown models built for local languages and governance needs. Bengaluru-based Sarvam AI has launched Sarvam Vision, an OCR system for complex paper records, and Bulbul, a text-to-speech model spanning 11 Indian languages. Sarvam claims its OCR beat Google Gemini on a benchmark. Backed by government funding, the programme aims to cut dependence on foreign AI systems. Watch this report. 🔽
5:05

India is accelerating its sovereign AI mission with homegrown models built for local languages and governance needs. Bengaluru-based Sarvam AI has launched Sarvam Vision, an OCR system for complex paper records, and Bulbul, a text-to-speech model spanning 11 Indian languages. Sarvam claims its OCR beat Google Gemini on a benchmark. Backed by government funding, the programme aims to cut dependence on foreign AI systems. Watch this report. 🔽

BJP

34,446 просмотров • 6 месяцев назад

Piddis ran campaign to undermine success of AI Revolution by #Modi Sarkar, but forgot that Bharat proved even Sam Altman wrong who said it's not possible to build LLM in Bharat from scratch with just millions in funding. Today SARVAM has launched its own foundational LLM, beating multiple foreign models on different parameters with just $53.4 Million funding till date. Sarvam AI has outperformed ChatGPT & Google Gemini in 3 key areas: 👉 OCR accuracy (84.3% on olmOCR-Bench) 👉 Indic languages & accents 👉 Voice-based AI (speech-to-text & text-to-speech) 😎AI Glasses: PM .Narendra Modi Ji became first person to try Sarvam Kaze build by Sarvam These Glasses can listen to any conversation and translate it into 22 different Indian Languages and you can listen to it as well. In May month, you can buy it on Shops or Online. Piddis want to undermine 'this' Bharat! I pity them.
0:44

Piddis ran campaign to undermine success of AI Revolution by #Modi Sarkar, but forgot that Bharat proved even Sam Altman wrong who said it's not possible to build LLM in Bharat from scratch with just millions in funding. Today SARVAM has launched its own foundational LLM, beating multiple foreign models on different parameters with just $53.4 Million funding till date. Sarvam AI has outperformed ChatGPT & Google Gemini in 3 key areas: 👉 OCR accuracy (84.3% on olmOCR-Bench) 👉 Indic languages & accents 👉 Voice-based AI (speech-to-text & text-to-speech) 😎AI Glasses: PM .Narendra Modi Ji became first person to try Sarvam Kaze build by Sarvam These Glasses can listen to any conversation and translate it into 22 different Indian Languages and you can listen to it as well. In May month, you can buy it on Shops or Online. Piddis want to undermine 'this' Bharat! I pity them.

BhikuMhatre

68,895 просмотров • 6 месяцев назад

Congratulations to Sarvam (Sarvam) on the first close of their Series B, and a warm welcome to HCLTech (HCLTech) and Bessemer Venture Partners India (Bessemer India), who join us in this journey with the conviction that has defined Sarvam from day one: that India should build and own its AI stack. In under three years, the team has turned that idea into working infrastructure: → Frontier models trained from scratch in India that hold their own against far larger systems → Speech models transcribing over half a million hours of audio a month in real Indian conditions → Voice agents already live, from banking and insurance to government and defence, reaching Indians at population scale| This is what sovereign AI looks like, and the team is just getting started. Here's Vivek on why sovereign AI matters, and why India can't afford to import its intelligence anymore. Once again, congratulations to Vivek (Vivek Raghavan), Pratyush (Pratyush Kumar), and the team. Hemant Mohapatra
0:41

Congratulations to Sarvam (Sarvam) on the first close of their Series B, and a warm welcome to HCLTech (HCLTech) and Bessemer Venture Partners India (Bessemer India), who join us in this journey with the conviction that has defined Sarvam from day one: that India should build and own its AI stack. In under three years, the team has turned that idea into working infrastructure: → Frontier models trained from scratch in India that hold their own against far larger systems → Speech models transcribing over half a million hours of audio a month in real Indian conditions → Voice agents already live, from banking and insurance to government and defence, reaching Indians at population scale| This is what sovereign AI looks like, and the team is just getting started. Here's Vivek on why sovereign AI matters, and why India can't afford to import its intelligence anymore. Once again, congratulations to Vivek (Vivek Raghavan), Pratyush (Pratyush Kumar), and the team. Hemant Mohapatra

Lightspeed India

12,153 просмотров • 2 месяцев назад

Everyone talks about "𝗔𝗜 𝗶𝗻 𝗜𝗻𝗱𝗶𝗮," but Sarvam AI just walked onto the stage at the India AI Impact Summit 2026 and showed the world what "𝗔𝗜 𝗯𝘆 𝗜𝗻𝗱𝗶𝗮" actually looks like. This is sovereign compute. 𝗧𝗵𝗲 𝗟𝗮𝘂𝗻𝗰𝗵: They didn't just launch one thing; they dropped an entire ecosystem tailored for 1.4 billion people. 𝗧𝗵𝗲 𝗛𝗲𝗮𝘃𝘆𝘄𝗲𝗶𝗴𝗵𝘁: Sarvam 105B & 30B 🧠 They unveiled two massive sovereign Large Language Models (LLMs) trained from scratch. 𝗦𝗮𝗿𝘃𝗮𝗺 𝟭𝟬𝟱𝗕:This is the beast. It’s a 105-billion parameter model that reportedly outperforms DeepSeek R1 on reasoning tasks and rivals global giants like Gemini Flash in efficiency. 𝗦𝗮𝗿𝘃𝗮𝗺 𝟯𝟬𝗕:The efficiency king, designed to run cost-effectively while handling complex Indic language reasoning. These aren't just translated models. They understand the nuance of 22 Indian languages, code-mixing (Hinglish, Tanglish), and cultural context that Western models often miss. Sarvam Kaze (Hardware!) 🕶️ This was the surprise "One More Thing" moment. ▶️They unveiled Sarvam Kaze, India’s first AI-powered smart glasses. ▶️PM Modi was the first person to demo them at the summit. ▶️They capture what you see and hear, processing it with their multimodal AI to give real-time intelligence. Launching May 2026. 𝗦𝗮𝗿𝘃𝗮𝗺 𝗔𝘂𝗱𝗶𝗼 & 𝗦𝗮𝗺𝘃𝗮𝗮𝗱 🗣️ An audio-first model that doesn't do "speech-to-text-to-LLM." It just hears and understands audio directly. It handles Indian accents, background noise, and interruptions flawlessly. 𝗛𝗼𝘄 𝗱𝗶𝗱 𝗮 𝘀𝘁𝗮𝗿𝘁𝘂𝗽 𝗮𝗰𝗵𝗶𝗲𝘃𝗲 𝘁𝗵𝗶𝘀? Building a 100B+ model isn't just about code; it's a logistics war. ▶️They secured 4,096 NVIDIA H100 GPUs (via Yotta Data Services). This is serious, nation-state level compute power. ▶️They trained on a massive 16 Trillion token dataset. Crucially, 2 Trillion of those were high-quality Indic tokens data that simply doesn't exist in the training sets of GPT-4 or Claude. ▶️They used a Mixture-of-Experts (MoE) architecture. This allows the model to be huge (smart) but only activate a fraction of parameters for each token (fast/cheap). ▶️They are a key part of the IndiaAI Mission, receiving subsidies and support to build "Sovereign AI" so India's data stays in India. 𝗪𝗵𝗼 𝗯𝘂𝗶𝗹𝘁 𝘁𝗵𝗶𝘀? 𝗣𝗿𝗮𝘁𝘆𝘂𝘀𝗵 𝗞𝘂𝗺𝗮𝗿 (𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿):The research heavyweight. Ex-IBM/Microsoft Research and IIT Bombay/Madras alum. He’s the one ensuring the models aren't just "big" but mathematically sound and efficient. 𝗩𝗶𝘃𝗲𝗸 𝗥𝗮𝗴𝗵𝗮𝘃𝗮𝗻 (𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿):The scale architect. He spent years with UIDAI (Aadhaar). He knows how to build systems that don't just work for a few thousand users, but for a billion people. For the last 3 years, the question was "𝗖𝗮𝗻 𝗜𝗻𝗱𝗶𝗮 𝗯𝘂𝗶𝗹𝗱 𝗮 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹?" Sarvam just answered: "Yes, and we can put it in hardware too." We are witnessing the shift from India being the "Back Office of the World" to the "Brain Office of the World." Col AJ🇮🇳 Major Sammer Pal Toorr (Infantry Combat Veteran) Navroop Singh Colonel Mayank Chaubey TheGlobalDecoder #SarvamAI #IndiaAIImpactSummit2026
0:44

Everyone talks about "𝗔𝗜 𝗶𝗻 𝗜𝗻𝗱𝗶𝗮," but Sarvam AI just walked onto the stage at the India AI Impact Summit 2026 and showed the world what "𝗔𝗜 𝗯𝘆 𝗜𝗻𝗱𝗶𝗮" actually looks like. This is sovereign compute. 𝗧𝗵𝗲 𝗟𝗮𝘂𝗻𝗰𝗵: They didn't just launch one thing; they dropped an entire ecosystem tailored for 1.4 billion people. 𝗧𝗵𝗲 𝗛𝗲𝗮𝘃𝘆𝘄𝗲𝗶𝗴𝗵𝘁: Sarvam 105B & 30B 🧠 They unveiled two massive sovereign Large Language Models (LLMs) trained from scratch. 𝗦𝗮𝗿𝘃𝗮𝗺 𝟭𝟬𝟱𝗕:This is the beast. It’s a 105-billion parameter model that reportedly outperforms DeepSeek R1 on reasoning tasks and rivals global giants like Gemini Flash in efficiency. 𝗦𝗮𝗿𝘃𝗮𝗺 𝟯𝟬𝗕:The efficiency king, designed to run cost-effectively while handling complex Indic language reasoning. These aren't just translated models. They understand the nuance of 22 Indian languages, code-mixing (Hinglish, Tanglish), and cultural context that Western models often miss. Sarvam Kaze (Hardware!) 🕶️ This was the surprise "One More Thing" moment. ▶️They unveiled Sarvam Kaze, India’s first AI-powered smart glasses. ▶️PM Modi was the first person to demo them at the summit. ▶️They capture what you see and hear, processing it with their multimodal AI to give real-time intelligence. Launching May 2026. 𝗦𝗮𝗿𝘃𝗮𝗺 𝗔𝘂𝗱𝗶𝗼 & 𝗦𝗮𝗺𝘃𝗮𝗮𝗱 🗣️ An audio-first model that doesn't do "speech-to-text-to-LLM." It just hears and understands audio directly. It handles Indian accents, background noise, and interruptions flawlessly. 𝗛𝗼𝘄 𝗱𝗶𝗱 𝗮 𝘀𝘁𝗮𝗿𝘁𝘂𝗽 𝗮𝗰𝗵𝗶𝗲𝘃𝗲 𝘁𝗵𝗶𝘀? Building a 100B+ model isn't just about code; it's a logistics war. ▶️They secured 4,096 NVIDIA H100 GPUs (via Yotta Data Services). This is serious, nation-state level compute power. ▶️They trained on a massive 16 Trillion token dataset. Crucially, 2 Trillion of those were high-quality Indic tokens data that simply doesn't exist in the training sets of GPT-4 or Claude. ▶️They used a Mixture-of-Experts (MoE) architecture. This allows the model to be huge (smart) but only activate a fraction of parameters for each token (fast/cheap). ▶️They are a key part of the IndiaAI Mission, receiving subsidies and support to build "Sovereign AI" so India's data stays in India. 𝗪𝗵𝗼 𝗯𝘂𝗶𝗹𝘁 𝘁𝗵𝗶𝘀? 𝗣𝗿𝗮𝘁𝘆𝘂𝘀𝗵 𝗞𝘂𝗺𝗮𝗿 (𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿):The research heavyweight. Ex-IBM/Microsoft Research and IIT Bombay/Madras alum. He’s the one ensuring the models aren't just "big" but mathematically sound and efficient. 𝗩𝗶𝘃𝗲𝗸 𝗥𝗮𝗴𝗵𝗮𝘃𝗮𝗻 (𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿):The scale architect. He spent years with UIDAI (Aadhaar). He knows how to build systems that don't just work for a few thousand users, but for a billion people. For the last 3 years, the question was "𝗖𝗮𝗻 𝗜𝗻𝗱𝗶𝗮 𝗯𝘂𝗶𝗹𝗱 𝗮 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹?" Sarvam just answered: "Yes, and we can put it in hardware too." We are witnessing the shift from India being the "Back Office of the World" to the "Brain Office of the World." Col AJ🇮🇳 Major Sammer Pal Toorr (Infantry Combat Veteran) Navroop Singh Colonel Mayank Chaubey TheGlobalDecoder #SarvamAI #IndiaAIImpactSummit2026

The Sacred Scroll

22,333 просмотров • 6 месяцев назад

Drop 11/14: We have been releasing Sarvam’s models and products - and yes, there is more to come. But today, we are excited to share the real and diverse impact our work is driving at scale. First up, preserving our cultural heritage. We have been working with Ekatra Foundation and Navajivan Trust (founded by Mahatma Gandhi) to digitize Gujarati documents from the early 19th and 20th centuries. Through this collaboration, we built out a whole product - Sarvam Akshar. Powered by the Sarvam Vision model, Akshar delivers state-of-the-art accuracy enabling reliable digitization of complex, real-world documents. Akshar sets the stage for our effort to leverage foundational models to preserve India’s cultural heritage - a direction we will double down on. Read more in our blog:
3:04

Drop 11/14: We have been releasing Sarvam’s models and products - and yes, there is more to come. But today, we are excited to share the real and diverse impact our work is driving at scale. First up, preserving our cultural heritage. We have been working with Ekatra Foundation and Navajivan Trust (founded by Mahatma Gandhi) to digitize Gujarati documents from the early 19th and 20th centuries. Through this collaboration, we built out a whole product - Sarvam Akshar. Powered by the Sarvam Vision model, Akshar delivers state-of-the-art accuracy enabling reliable digitization of complex, real-world documents. Akshar sets the stage for our effort to leverage foundational models to preserve India’s cultural heritage - a direction we will double down on. Read more in our blog:

Pratyush Kumar

60,903 просмотров • 6 месяцев назад

India AI Impact Summit is about many stories on indigenous innovations and achievements. It is a marquee event for display of present and future AI technologies at international level. Each such story is a testament to India's growing AI ecosystem. There are hundreds of stalls displaying futuristic tech being developed in India. From ground zero, I covered the innovations shaping India's AI future. BharatGen AI: Building foundational models in all 22 official Indian languages. ParadigmIT's Sovereign AI Box: Strengthening data security and aligning with the IndiaAI Mission. Skye Air Mobility: 3.6 million autonomous deliveries and 1,000+ tonnes of CO₂ saved. I brought forward stories that matter: Tarakram Maram's AI Trainer Machine, democratizing AI education. Frontier Markets which is empowering rural women entrepreneurs through AI. Drublet Innovation Private Limited, Founded by two students, Agniva and Aaditya, pushing the boundaries of autonomous navigation and robotics. I also explored state pavilions like Bihar's, interacted with international delegations, and highlighted how startups across India are solving real, local challenges with AI. As a journalist on the ground, my focus was simple: bringing the stories of innovation, impact, and aspiration to the viewers. The AI Summit also displays various budding innovations from startups like Sarvam to presence of Indian tech giants like HCL, Airtel, Tata. Top names in the field of AI from across the world like an NVIDIA, Google, OpenAI and many more have also lined up to participate in the Summit. One story about Galgotia University which seemed to go wrong is just an isolated droplet of an otherwise vast reservoir of innovation and future ready technology. If 1 out of hundreds of exhibitors wasn’t being upfront about their innovation, I would not give up on the entire India’s youth who are very innovative. Would you also trash the entire India story just because of one episode? For all tech enthusiasts and those interested in the future of AI, the Summit is a vibrant opportunity to understand how India is at the forefront of the AI revolution. Do visit! #IndiaAISummit2026 #IndiaAIMegaEvent #IndiaForAI #IndiaAIImpactSummit2026 #IndiaAISummit2026 #IndiaAIExpo #ViksitBharat
2:52

India AI Impact Summit is about many stories on indigenous innovations and achievements. It is a marquee event for display of present and future AI technologies at international level. Each such story is a testament to India's growing AI ecosystem. There are hundreds of stalls displaying futuristic tech being developed in India. From ground zero, I covered the innovations shaping India's AI future. BharatGen AI: Building foundational models in all 22 official Indian languages. ParadigmIT's Sovereign AI Box: Strengthening data security and aligning with the IndiaAI Mission. Skye Air Mobility: 3.6 million autonomous deliveries and 1,000+ tonnes of CO₂ saved. I brought forward stories that matter: Tarakram Maram's AI Trainer Machine, democratizing AI education. Frontier Markets which is empowering rural women entrepreneurs through AI. Drublet Innovation Private Limited, Founded by two students, Agniva and Aaditya, pushing the boundaries of autonomous navigation and robotics. I also explored state pavilions like Bihar's, interacted with international delegations, and highlighted how startups across India are solving real, local challenges with AI. As a journalist on the ground, my focus was simple: bringing the stories of innovation, impact, and aspiration to the viewers. The AI Summit also displays various budding innovations from startups like Sarvam to presence of Indian tech giants like HCL, Airtel, Tata. Top names in the field of AI from across the world like an NVIDIA, Google, OpenAI and many more have also lined up to participate in the Summit. One story about Galgotia University which seemed to go wrong is just an isolated droplet of an otherwise vast reservoir of innovation and future ready technology. If 1 out of hundreds of exhibitors wasn’t being upfront about their innovation, I would not give up on the entire India’s youth who are very innovative. Would you also trash the entire India story just because of one episode? For all tech enthusiasts and those interested in the future of AI, the Summit is a vibrant opportunity to understand how India is at the forefront of the AI revolution. Do visit! #IndiaAISummit2026 #IndiaAIMegaEvent #IndiaForAI #IndiaAIImpactSummit2026 #IndiaAISummit2026 #IndiaAIExpo #ViksitBharat

Tapas Bhattachary

25,109 просмотров • 6 месяцев назад

The teams shipping AI agents right now are bleeding money on the dumbest possible expense: teaching a 400B-parameter model to read a file name. Every time an AI agent needs to "see" something today, it routes an image through a frontier model. OCR, object detection, checking if a button exists on screen. You're paying GPT-4o or Claude pricing for tasks that require perception, not reasoning. One agent workflow processing a few thousand screenshots per day can burn through more on vision calls than on the actual thinking. Perceptron's Isaac is 2B parameters. Built by the team that created Meta's Chameleon multimodal models. On perceptive benchmarks, it matches or beats models 50x its size. The VQA, OCR, and object detection scores are competitive with models running on infrastructure that costs orders of magnitude more. The MCP wrapper is the distribution play. One install command and every Claude Code agent can offload vision tasks to a model that runs on a single consumer GPU. The agent keeps its reasoning in the frontier model and routes perception to a specialist. That split is how you get vision-heavy agent workflows from "technically possible but expensive" to "cheap enough to run on everything." This is the same pattern that won in every other compute-intensive stack. General-purpose handles orchestration. Specialists handle the heavy lifting. Graphics went through it. Audio went through it. Video encoding went through it. Vision in AI agents is next. The teams building agents that see 10,000 images a day will care about this before anyone else does.
0:38

The teams shipping AI agents right now are bleeding money on the dumbest possible expense: teaching a 400B-parameter model to read a file name. Every time an AI agent needs to "see" something today, it routes an image through a frontier model. OCR, object detection, checking if a button exists on screen. You're paying GPT-4o or Claude pricing for tasks that require perception, not reasoning. One agent workflow processing a few thousand screenshots per day can burn through more on vision calls than on the actual thinking. Perceptron's Isaac is 2B parameters. Built by the team that created Meta's Chameleon multimodal models. On perceptive benchmarks, it matches or beats models 50x its size. The VQA, OCR, and object detection scores are competitive with models running on infrastructure that costs orders of magnitude more. The MCP wrapper is the distribution play. One install command and every Claude Code agent can offload vision tasks to a model that runs on a single consumer GPU. The agent keeps its reasoning in the frontier model and routes perception to a specialist. That split is how you get vision-heavy agent workflows from "technically possible but expensive" to "cheap enough to run on everything." This is the same pattern that won in every other compute-intensive stack. General-purpose handles orchestration. Specialists handle the heavy lifting. Graphics went through it. Audio went through it. Video encoding went through it. Vision in AI agents is next. The teams building agents that see 10,000 images a day will care about this before anyone else does.

Aakash Gupta

55,978 просмотров • 5 месяцев назад

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:
0:44

This is THE moment of Physical AI! We are officially announcing Cosmos 3: Omnimodal World Models for Physical AI 🚀 - Cosmos 3 is an omnimodal world model: within a unified architecture, it can understand and generate language, images, video, audio, and actions. - It is not just a VLM, not just a video generator, not just an audio-visual generative model, and not just a physics simulator / world-action model. It can understand images and videos, generate images, videos, and audio, simulate future worlds, predict actions, and generate robot policies—enabling models to truly begin to “touch the world.” - Cosmos 3 is the #1 open-weight reasoner / T2I / I2V / robot policy across many benchmarks. Huge thanks to every teammate who fought side by side on this journey—from architecture, data, training, infra, serving, and evaluation to post-training. Every part of this project carries an incredible amount of hard work. This was my first time leading a project as Tech Lead, and I feel truly fortunate. The future of Physical AI needs models that can not only “see” and “describe” the world, but also “imagine,” “simulate,” and “act”—and eventually close the loop with the real world. I hope Cosmos 3 can become an important starting point for this direction, and I’m excited to push Physical AI into its next stage together with the open-source community. Welcome to the era of Physical AI. HuggingFace: Project Website: Code:

Max Zhaoshuo Li 李赵硕

1,078,810 просмотров • 3 месяцев назад

Bihar is taking important steps toward building an AI-driven future. The Bihar Ai Summit brought together government leaders, policymakers, industry representatives, academia, and technology experts to discuss how Artificial Intelligence can accelerate innovation, improve public service delivery, and create new opportunities across sectors. The presence of Hon'ble Chief Minister Shri Samrat Choudhary Ji, Hon'ble Deputy Chief Minister Shri Bijendra Prasad Yadav Ji, Hon'ble IT Minister Shri Nitish Mishra Ji, IT Secretary Shri Abhay Kumar Singh Ji, Hon'ble DGP Mr Vinay Kumar ji, IPS and other distinguished participants reflects Bihar's commitment to leveraging technology for economic growth and digital transformation. As AI adoption expands across governance, healthcare, education, agriculture, banking, and enterprise, the focus must remain on building practical, accessible, and impactful applications that address real-world challenges. At CoRover | BharatGPT, we believe the next phase of AI will be driven by multilingual, inclusive, and population-scale innovation built on local knowledge, local languages, and local expertise. Bihar's young talent base, academic institutions, and growing technology ecosystem create a strong foundation for this journey. Sharing highlights from the Bihar AI Summit, I spoke about the opportunity to build AI in Indian languages, enable innovation beyond major technology hubs, and create technologies that can serve both India and the world. India's AI growth story will be strengthened when more states, institutions, entrepreneurs, and innovators contribute to building solutions for India and the world. Full video: #AI #SovereignAI #BharatGPT #IndiaAI #AgenticAI #AIInnovation #ResponsibleAI #DigitalTransformation
10:26

Bihar is taking important steps toward building an AI-driven future. The Bihar Ai Summit brought together government leaders, policymakers, industry representatives, academia, and technology experts to discuss how Artificial Intelligence can accelerate innovation, improve public service delivery, and create new opportunities across sectors. The presence of Hon'ble Chief Minister Shri Samrat Choudhary Ji, Hon'ble Deputy Chief Minister Shri Bijendra Prasad Yadav Ji, Hon'ble IT Minister Shri Nitish Mishra Ji, IT Secretary Shri Abhay Kumar Singh Ji, Hon'ble DGP Mr Vinay Kumar ji, IPS and other distinguished participants reflects Bihar's commitment to leveraging technology for economic growth and digital transformation. As AI adoption expands across governance, healthcare, education, agriculture, banking, and enterprise, the focus must remain on building practical, accessible, and impactful applications that address real-world challenges. At CoRover | BharatGPT, we believe the next phase of AI will be driven by multilingual, inclusive, and population-scale innovation built on local knowledge, local languages, and local expertise. Bihar's young talent base, academic institutions, and growing technology ecosystem create a strong foundation for this journey. Sharing highlights from the Bihar AI Summit, I spoke about the opportunity to build AI in Indian languages, enable innovation beyond major technology hubs, and create technologies that can serve both India and the world. India's AI growth story will be strengthened when more states, institutions, entrepreneurs, and innovators contribute to building solutions for India and the world. Full video: #AI #SovereignAI #BharatGPT #IndiaAI #AgenticAI #AIInnovation #ResponsibleAI #DigitalTransformation

Ankush Sabharwal

165,688 просмотров • 3 месяцев назад

Exclusive: Meta just released Llama 3.1 405B — the first-ever open-sourced frontier AI model, beating top closed models like GPT-4o across several benchmarks. I sat down with Mark Zuckerberg, diving into why this marks a major moment in AI history. Timestamps: 00:00 Intro 00:38 Meta’s Llama 3.1 rundown 03:44 Real-world use cases for Llama 3.1 06:15 Educating developers on open-source AI tools 09:43 Societal implications of open-source AI 13:00 Balancing power and managing bad actors 14:40 Open source and global competition 16:59 Accelerating innovation and economic growth 20:04 Zuck on Apple and lessons from the past 24:22 Future of AI: Llama 3 and beyond 26:43 Prediction: Billions of personalized AI agents 31:32 Factors to changing anti-AI sentiment
35:22

Exclusive: Meta just released Llama 3.1 405B — the first-ever open-sourced frontier AI model, beating top closed models like GPT-4o across several benchmarks. I sat down with Mark Zuckerberg, diving into why this marks a major moment in AI history. Timestamps: 00:00 Intro 00:38 Meta’s Llama 3.1 rundown 03:44 Real-world use cases for Llama 3.1 06:15 Educating developers on open-source AI tools 09:43 Societal implications of open-source AI 13:00 Balancing power and managing bad actors 14:40 Open source and global competition 16:59 Accelerating innovation and economic growth 20:04 Zuck on Apple and lessons from the past 24:22 Future of AI: Llama 3 and beyond 26:43 Prediction: Billions of personalized AI agents 31:32 Factors to changing anti-AI sentiment

Rowan Cheung

2,657,507 просмотров • 2 лет назад

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️
1:36

Open science is how we continue to push technology forward and today at Meta FAIR we’re sharing eight new AI research artifacts including new models, datasets and code to inspire innovation in the community. More in the video from Joelle Pineau. This work is another important step towards our goal of achieving Advanced Machine Intelligence (AMI). What we’re releasing: • Meta Spirit LM: An open source language model for seamless speech and text integration. • Meta Segment Anything Model 2.1: An updated checkpoint with improved results on visually similar objects, small objects and occlusion handling. Plus a new developer suite to make it easier for developers to build with SAM 2. • Layer Skip: Inference code and fine-tuned checkpoints demonstrating a new method for enhancing LLM performance. • SALSA: New code to enable researchers to benchmark AI-based attacks in support of validating security for post-quantum cryptography. • Meta Lingua: A lightweight and self-contained codebase designed to train language models at scale. • Meta Open Materials: New open source models and the largest dataset of its kind to accelerate AI-driven discovery of new inorganic materials. • MEXMA: A new research paper and code for our novel pre-trained cross-lingual sentence encoder with coverage across 80 languages. • Self-Taught Evaluator: a new method for generating synthetic preference data to train reward models without relying on human annotations. Access to state-of-the-art AI creates opportunities for everyone. We’re excited to share this work and look forward to seeing the community innovation that results from it. Details and access to everything released by FAIR today ➡️

AI at Meta

150,435 просмотров • 1 год назад

DeepSeek-R1 shattered the assumption that performant AI models must be built closed source with loss-leading computational costs. This is the reality that Web3 x Crypto firms have been waiting for, leading me to believe that the most performant AI models in the future will be built on-chain. Resource Requirements DeepSeek R1 (671 billion parameters), which took over a billion dollars, 2,000 Nvidia H800 GPUs, and over 55 days, beat benchmarks held by OpenAI’s o1 mode (near 2 trillion parameters)l, which required hundreds of billions of dollars to develop along with over 16,000 advanced GPUs. The idea that AI models must be closed-source and have loss-leading computational costs to succeed is crumbling. The Existing Decentralized AI Narrative AI x Crypto projects believed that crowdsourced, public, decentralized AI would eventually create better models than their centralized counterparts. This had thus far not been true, as the highest-performing models had come from closed-source companies like OpenAI and Anthropic. Crypto x AI companies have adapted to this by specializing in infrastructure rather than model-building. For example, GPU marketplaces like , The Render Network, io.net, and Exabits have developed sustainable revenues. Companies that allow users to share their network bandwidth like touch grass and Gradient have found their niche in supplying services, like distributed web scraping, to web2 clients. Storage networks like Arweave Ecosystem, Filecoin, and Ocean Protocol have also done well by being the platform on which these projects are built. Supply networks have flourished because of their ability to tailor their cheaper and more scalable services to off-chain customers. Renewed Focus Now that GPU and financial resources are no longer limitations to creating quality AI models, web3 AI companies can focus on replicating DeepSeek’s effectiveness while offering new benefits like modality, user ownership, censorship resistance, privacy, and more. Pantera Capital has funded companies in this space like and Sentient that believe they can match or exceed the performance of traditional AI companies while offering additional services or benefits. , for example, is building a platform where anyone can monetize AI models, data sets, and applications in a collaborative space. Users can permissionlessly train models manually, provide training data, and create tailored AI models with no-code tools. They are only able to cater to all these stakeholders (AI developers, users, resource providers) because everything is tied to their native Sahara blockchain. We invested in them precisely for this reason. The Future of AI will be built with Web3 Infrastructure I believe that supply-side projects will continue to grow, while consumer-facing projects can begin competing with web2 competitors by taking advantage of their ability to build networks that invite community involvement. and Sentient, for example, have begun setting up systems for users to train models based on the users’ expertise. These platforms will allow users to pick and choose the data and integrations to whatever they are applying the model towards. Sahara already has over 780,000 users on their waitlist while Sentient has over 1 million interactions. In the near future, I believe that the most performant AI models will be built on-chain. For the full blog post, read my newsletter.
0:32

DeepSeek-R1 shattered the assumption that performant AI models must be built closed source with loss-leading computational costs. This is the reality that Web3 x Crypto firms have been waiting for, leading me to believe that the most performant AI models in the future will be built on-chain. Resource Requirements DeepSeek R1 (671 billion parameters), which took over a billion dollars, 2,000 Nvidia H800 GPUs, and over 55 days, beat benchmarks held by OpenAI’s o1 mode (near 2 trillion parameters)l, which required hundreds of billions of dollars to develop along with over 16,000 advanced GPUs. The idea that AI models must be closed-source and have loss-leading computational costs to succeed is crumbling. The Existing Decentralized AI Narrative AI x Crypto projects believed that crowdsourced, public, decentralized AI would eventually create better models than their centralized counterparts. This had thus far not been true, as the highest-performing models had come from closed-source companies like OpenAI and Anthropic. Crypto x AI companies have adapted to this by specializing in infrastructure rather than model-building. For example, GPU marketplaces like , The Render Network, io.net, and Exabits have developed sustainable revenues. Companies that allow users to share their network bandwidth like touch grass and Gradient have found their niche in supplying services, like distributed web scraping, to web2 clients. Storage networks like Arweave Ecosystem, Filecoin, and Ocean Protocol have also done well by being the platform on which these projects are built. Supply networks have flourished because of their ability to tailor their cheaper and more scalable services to off-chain customers. Renewed Focus Now that GPU and financial resources are no longer limitations to creating quality AI models, web3 AI companies can focus on replicating DeepSeek’s effectiveness while offering new benefits like modality, user ownership, censorship resistance, privacy, and more. Pantera Capital has funded companies in this space like and Sentient that believe they can match or exceed the performance of traditional AI companies while offering additional services or benefits. , for example, is building a platform where anyone can monetize AI models, data sets, and applications in a collaborative space. Users can permissionlessly train models manually, provide training data, and create tailored AI models with no-code tools. They are only able to cater to all these stakeholders (AI developers, users, resource providers) because everything is tied to their native Sahara blockchain. We invested in them precisely for this reason. The Future of AI will be built with Web3 Infrastructure I believe that supply-side projects will continue to grow, while consumer-facing projects can begin competing with web2 competitors by taking advantage of their ability to build networks that invite community involvement. and Sentient, for example, have begun setting up systems for users to train models based on the users’ expertise. These platforms will allow users to pick and choose the data and integrations to whatever they are applying the model towards. Sahara already has over 780,000 users on their waitlist while Sentient has over 1 million interactions. In the near future, I believe that the most performant AI models will be built on-chain. For the full blog post, read my newsletter.

paul.nft

32,465 просмотров • 1 год назад

U.N. official wants to decolonize AI and train new models with third-world "systems of knowledge" Deputy Secretary-General Amina J. Mohammed of Nigeria: "Colonial conquest brought catastrophe and genocide, in many cases on a continental scale. Entire communities were destroyed. Indigenous people were driven far from their lands. The assault was also on memory itself. Languages were suppressed. Knowledge gained through millennia of civilization, including in the Mayan codices, was deliberately put to the flame. Many traditions rooted in land and community and memory were lost." "These are living traditions carrying a wisdom that can help the region shape technological change in its own image towards its own vision of a good life, to pursue development on its own terms. And that starts with ensuring the people of this region can see themselves in the technology. More than 800 indigenous peoples live here, representing 60 million people. Many of their languages and systems of knowledge are sustained through oral traditions when AI systems are trained largely on written material that has been gathered elsewhere. That knowledge can be distorted or made invisible. Right now, Chile's National Center for Artificial Intelligence and 30 partner institutions are developing a LatAm GPT, an open-source model trained on regional data. Its first version is in Spanish and Portuguese, but researchers are also developing tools for other languages -- with the consent of the indigenous communities concerned. And that is what an inclusive reset can look like. Technology built with a fuller account of the people that it is meant to serve."
2:19

U.N. official wants to decolonize AI and train new models with third-world "systems of knowledge" Deputy Secretary-General Amina J. Mohammed of Nigeria: "Colonial conquest brought catastrophe and genocide, in many cases on a continental scale. Entire communities were destroyed. Indigenous people were driven far from their lands. The assault was also on memory itself. Languages were suppressed. Knowledge gained through millennia of civilization, including in the Mayan codices, was deliberately put to the flame. Many traditions rooted in land and community and memory were lost." "These are living traditions carrying a wisdom that can help the region shape technological change in its own image towards its own vision of a good life, to pursue development on its own terms. And that starts with ensuring the people of this region can see themselves in the technology. More than 800 indigenous peoples live here, representing 60 million people. Many of their languages and systems of knowledge are sustained through oral traditions when AI systems are trained largely on written material that has been gathered elsewhere. That knowledge can be distorted or made invisible. Right now, Chile's National Center for Artificial Intelligence and 30 partner institutions are developing a LatAm GPT, an open-source model trained on regional data. Its first version is in Spanish and Portuguese, but researchers are also developing tools for other languages -- with the consent of the indigenous communities concerned. And that is what an inclusive reset can look like. Technology built with a fuller account of the people that it is meant to serve."

Breitbart News

45,256 просмотров • 1 месяц назад

Orbit AI Satellite Successfully Achieve World’s First Orbital AI Deployment and Launching Digital AI Sovereignty Decentralized Orbital AI Network Orbit AI Orbit AI🛰️ today announced that the first satellite, “OAI Genesis-1,” has successfully launched and entered Low Earth Orbit (LEO). Amidst fierce competition from tech giants (e.g., Starlink Starlink Elon Musk , Google AI Project Suncatcher) in space AI computing, this launch signifies Orbit AI’s position as the first to achieve real-world AI deployment, formally inaugurating its "Orbit AI Cloud Platform." Genesis-1 is equipped with NVIDIA NVIDIA AI Compute Cores, running a 2.6B parameter AI model for real-time analysis of infrared remote sensing data in space. By processing data on orbit, Genesis-1 drastically reduces critical information retrieval time (e.g., disaster alerts, maritime monitoring) from hours to mere seconds, while cutting transmission bandwidth costs by over 90%. Furthermore, Orbit AI has partnered with from energy company Powerbank (NASDAQ: SUUN) ( utilizing infinite solar power to achieve carbon-neutral computing and projecting a reduction in overall energy operational costs by 60%. Following its triumph at the BNB Chain Hackathon ( Orbit AI protocol is committed to creating an ultimate censorship-resistant deployment environment: Developers can deploy AI models, privacy applications, financial algorithms, and even blockchain nodes on the satellite network. This ensures that code and data operate in a physically isolated, neutral environment beyond the jurisdiction of major nations, guaranteeing extreme digital sovereignty and service resilience. Orbit AI will also leverage the RWA (Real World Assets) mechanism to allow community users to purchase satellite NFT shares, becoming co-owners of this space infrastructure and sharing in its compute revenues, thus building a community-owned orbital AI economy.
0:20

Orbit AI Satellite Successfully Achieve World’s First Orbital AI Deployment and Launching Digital AI Sovereignty Decentralized Orbital AI Network Orbit AI Orbit AI🛰️ today announced that the first satellite, “OAI Genesis-1,” has successfully launched and entered Low Earth Orbit (LEO). Amidst fierce competition from tech giants (e.g., Starlink Starlink Elon Musk , Google AI Project Suncatcher) in space AI computing, this launch signifies Orbit AI’s position as the first to achieve real-world AI deployment, formally inaugurating its "Orbit AI Cloud Platform." Genesis-1 is equipped with NVIDIA NVIDIA AI Compute Cores, running a 2.6B parameter AI model for real-time analysis of infrared remote sensing data in space. By processing data on orbit, Genesis-1 drastically reduces critical information retrieval time (e.g., disaster alerts, maritime monitoring) from hours to mere seconds, while cutting transmission bandwidth costs by over 90%. Furthermore, Orbit AI has partnered with from energy company Powerbank (NASDAQ: SUUN) ( utilizing infinite solar power to achieve carbon-neutral computing and projecting a reduction in overall energy operational costs by 60%. Following its triumph at the BNB Chain Hackathon ( Orbit AI protocol is committed to creating an ultimate censorship-resistant deployment environment: Developers can deploy AI models, privacy applications, financial algorithms, and even blockchain nodes on the satellite network. This ensures that code and data operate in a physically isolated, neutral environment beyond the jurisdiction of major nations, guaranteeing extreme digital sovereignty and service resilience. Orbit AI will also leverage the RWA (Real World Assets) mechanism to allow community users to purchase satellite NFT shares, becoming co-owners of this space infrastructure and sharing in its compute revenues, thus building a community-owned orbital AI economy.

Orbit AI🛰️

24,839 просмотров • 8 месяцев назад

I am stocked to announce that I won the OpenAI Developers Codex x Mollie Hacka Worldwide Hackathon in Paris. 60+ builders, every one of us working solo, one day to ship. I built mine around a single question: who gets to own intelligence? The default answer is scary. You hand your data to a handful of labs, they train the model, they own it, and you rent back a thin slice of what your own data made possible. That is the bargain on the table today. I do not accept it. So I built Lensemble: a Tapestry like distributed training platform for JEPA based World Models. What does it enable: World Models that a community improves together, keeps sovereign, and co-owns. Two bets sit underneath it. First, the paradigm. Language models predict the next token. Powerful for text, a dead end for the physical world. A robot does not need to autocomplete sentences, it needs to predict what happens next in the world. That is what JEPA does: it learns by predicting representations instead of pixels or tokens. I am convinced world models are the most underrated paradigm in AI right now, and the closest thing we have to a ChatGPT moment for robotics. Second, the politics. Your raw trajectories never leave your machine. Each participant trains locally against a shared protocol and ships only an update, never the data. A federated round folds those updates into one shared world model, a LeWorldModel based model, and the gain is measured, not claimed: a 12k-parameter adapter on a frozen backbone, held-out prediction error down about 12 percent, the model measurably less surprised by the world. Then the upside is split by contribution weight, so the people who improved the model own a share of what it earns. This is the thesis behind Project Tapestry, the AI Alliance and Yann LeCun's push for federated, sovereign frontier AI, carried into world models and robotics. Call it Tapestry for the physical world. All of it built solo, in a single day, with Codex as my pair the whole way. Thank you to OpenAI Codex and Mollie for backing builders who ship real things, and to Boris and the organizing crew for the room and the standard you set. Intelligence the world improves, and the world owns. That is the future I want for my kids, and the one I will keep building.
1:13

I am stocked to announce that I won the OpenAI Developers Codex x Mollie Hacka Worldwide Hackathon in Paris. 60+ builders, every one of us working solo, one day to ship. I built mine around a single question: who gets to own intelligence? The default answer is scary. You hand your data to a handful of labs, they train the model, they own it, and you rent back a thin slice of what your own data made possible. That is the bargain on the table today. I do not accept it. So I built Lensemble: a Tapestry like distributed training platform for JEPA based World Models. What does it enable: World Models that a community improves together, keeps sovereign, and co-owns. Two bets sit underneath it. First, the paradigm. Language models predict the next token. Powerful for text, a dead end for the physical world. A robot does not need to autocomplete sentences, it needs to predict what happens next in the world. That is what JEPA does: it learns by predicting representations instead of pixels or tokens. I am convinced world models are the most underrated paradigm in AI right now, and the closest thing we have to a ChatGPT moment for robotics. Second, the politics. Your raw trajectories never leave your machine. Each participant trains locally against a shared protocol and ships only an update, never the data. A federated round folds those updates into one shared world model, a LeWorldModel based model, and the gain is measured, not claimed: a 12k-parameter adapter on a frozen backbone, held-out prediction error down about 12 percent, the model measurably less surprised by the world. Then the upside is split by contribution weight, so the people who improved the model own a share of what it earns. This is the thesis behind Project Tapestry, the AI Alliance and Yann LeCun's push for federated, sovereign frontier AI, carried into world models and robotics. Call it Tapestry for the physical world. All of it built solo, in a single day, with Codex as my pair the whole way. Thank you to OpenAI Codex and Mollie for backing builders who ship real things, and to Boris and the organizing crew for the room and the standard you set. Intelligence the world improves, and the world owns. That is the future I want for my kids, and the one I will keep building.

abdel

20,037 просмотров • 2 месяцев назад

Honoured to participate in the CNBC-TV18 Global AI Lens Fireside Conversation at the #IndiaAIImpactSummit2026, following insightful remarks by H.E. Ebba Busch, Deputy Prime Minister of Sweden, where I reflected on how the India–Sweden partnership combines Sweden’s world-class innovation with India’s unparalleled scale as a real-world test bed for AI deployment. I emphasised that while the Union Government provides national frameworks and digital public infrastructure, the real momentum of AI adoption will be driven by states. Therefore, competitive and collaborative federalism must become the engine of implementation, and closer Centre–State coordination is essential to translate policy ambition into measurable outcomes for citizens. Tamil Nadu is leading this charge as an enabler by strengthening structured data systems, offering calibrated incentives, and deploying practical AI solutions in high-impact sectors such as health, agriculture, and governance. Inclusion is not optional; accordingly, we are expanding 100 Mbps fiber connectivity to every village and providing AI-enabled laptops to college students to ensure our youth are prepared for the future. To ensure true social equity, we must move beyond English and text-based interfaces toward voice-based and local language models that serve everyone, regardless of literacy. By balancing innovation with responsibility and equity, we are positioning India as a trusted and inclusive partner in the global AI ecosystem. Watch the video here : [English W\ Tamil CC]
11:34

Honoured to participate in the CNBC-TV18 Global AI Lens Fireside Conversation at the #IndiaAIImpactSummit2026, following insightful remarks by H.E. Ebba Busch, Deputy Prime Minister of Sweden, where I reflected on how the India–Sweden partnership combines Sweden’s world-class innovation with India’s unparalleled scale as a real-world test bed for AI deployment. I emphasised that while the Union Government provides national frameworks and digital public infrastructure, the real momentum of AI adoption will be driven by states. Therefore, competitive and collaborative federalism must become the engine of implementation, and closer Centre–State coordination is essential to translate policy ambition into measurable outcomes for citizens. Tamil Nadu is leading this charge as an enabler by strengthening structured data systems, offering calibrated incentives, and deploying practical AI solutions in high-impact sectors such as health, agriculture, and governance. Inclusion is not optional; accordingly, we are expanding 100 Mbps fiber connectivity to every village and providing AI-enabled laptops to college students to ensure our youth are prepared for the future. To ensure true social equity, we must move beyond English and text-based interfaces toward voice-based and local language models that serve everyone, regardless of literacy. By balancing innovation with responsibility and equity, we are positioning India as a trusted and inclusive partner in the global AI ecosystem. Watch the video here : [English W\ Tamil CC]

Dr P Thiaga Rajan (PTR)

22,196 просмотров • 6 месяцев назад

Reinforcement Learning from Human Feedback (RLHF) is gaining traction. This field aims to make AI more responsible by including human values and preferences. In this video, Nathan Lambert, a research scientist and RLHF team lead at Hugging Face explores its inner workings, applications and industry impact. RLHF has gained the spotlight in recent years. The growth of language models like Anthropic’s Claude and OpenAI's ChatGPT have increased interest in human-feedback integration. "There are some rumors that Open AI had two teams; one was doing RLHF and the other instruction fine-tuning. And the RLHF team kept getting more and more performance." Understanding RLHF The RLHF process has three main steps: Pre-training: Much like with GPT models, the journey starts with pre-training on a large corpus of data. This can range from text data, web scrapes, to specialized datasets. Reward Modeling: This is the RLHF counterpart of supervised fine-tuning in large language models. This stage involves creating a reward model that resonates with human values and preferences. RL Optimization: This stage parallels reward modeling and reinforcement learning in traditional AI models. The AI system fine-tunes itself based on the reward model, employing reinforcement learning algorithms for that extra layer of optimization. The Data Challenge Data collection and curation in RLHF closely resemble the challenges you'd encounter in large language model training. Datasets from organizations like OpenAI can serve as a useful foundation. However, the need for high-quality, task-specific data cannot be overstated. Implementing RLHF: A Practical Guide If you’re someone who loves getting hands-on with AI libraries like Hugging Face, implementing RLHF is right way to do. It’s essential to understand its limitations. Think about model stability, over-optimization, and exploration strategies, much like you would when prompt engineering. Ongoing Research and Next Steps While he suggests that some basics figured out, there are layers of complexity that still need to be unraveled: 1. New Benchmarks: How do we measure the effectiveness of RLHF? 2. Preference Modeling: How can the model be made to understand human preferences better? 3. Interpreting RLHF: Much like explainability in traditional models, how do we make RLHF more interpretable? 4. System-Wide Evaluation: Going beyond individual performance, how does RLHF affect an entire system? The Transformative Power of RLHF Whether you're an AI developer, a business analyst, or a marketer, RLHF promises to revolutionize your domain. Imagine customer service chatbots that understand human emotions better, or content generators that align more closely with human values. RLHF is an emerging field that focuses on enhancing machine learning models through human feedback. While it tackles important issues like bias and ethics, its broader goal is to improve system performance across various applications. Whether you're deeply invested in the ethics of AI or simply curious about advancements in machine learning, RLHF offers valuable insights. If you're interested in the next wave of AI development, this area is definitely one to watch.
2:00

Reinforcement Learning from Human Feedback (RLHF) is gaining traction. This field aims to make AI more responsible by including human values and preferences. In this video, Nathan Lambert, a research scientist and RLHF team lead at Hugging Face explores its inner workings, applications and industry impact. RLHF has gained the spotlight in recent years. The growth of language models like Anthropic’s Claude and OpenAI's ChatGPT have increased interest in human-feedback integration. "There are some rumors that Open AI had two teams; one was doing RLHF and the other instruction fine-tuning. And the RLHF team kept getting more and more performance." Understanding RLHF The RLHF process has three main steps: Pre-training: Much like with GPT models, the journey starts with pre-training on a large corpus of data. This can range from text data, web scrapes, to specialized datasets. Reward Modeling: This is the RLHF counterpart of supervised fine-tuning in large language models. This stage involves creating a reward model that resonates with human values and preferences. RL Optimization: This stage parallels reward modeling and reinforcement learning in traditional AI models. The AI system fine-tunes itself based on the reward model, employing reinforcement learning algorithms for that extra layer of optimization. The Data Challenge Data collection and curation in RLHF closely resemble the challenges you'd encounter in large language model training. Datasets from organizations like OpenAI can serve as a useful foundation. However, the need for high-quality, task-specific data cannot be overstated. Implementing RLHF: A Practical Guide If you’re someone who loves getting hands-on with AI libraries like Hugging Face, implementing RLHF is right way to do. It’s essential to understand its limitations. Think about model stability, over-optimization, and exploration strategies, much like you would when prompt engineering. Ongoing Research and Next Steps While he suggests that some basics figured out, there are layers of complexity that still need to be unraveled: 1. New Benchmarks: How do we measure the effectiveness of RLHF? 2. Preference Modeling: How can the model be made to understand human preferences better? 3. Interpreting RLHF: Much like explainability in traditional models, how do we make RLHF more interpretable? 4. System-Wide Evaluation: Going beyond individual performance, how does RLHF affect an entire system? The Transformative Power of RLHF Whether you're an AI developer, a business analyst, or a marketer, RLHF promises to revolutionize your domain. Imagine customer service chatbots that understand human emotions better, or content generators that align more closely with human values. RLHF is an emerging field that focuses on enhancing machine learning models through human feedback. While it tackles important issues like bias and ethics, its broader goal is to improve system performance across various applications. Whether you're deeply invested in the ethics of AI or simply curious about advancements in machine learning, RLHF offers valuable insights. If you're interested in the next wave of AI development, this area is definitely one to watch.

Muratcan Koylan

27,168 просмотров • 2 лет назад

AI is changing sports. Here is how. I sit down with Max Sebti, , founder and CEO of Score, and he gives me the latest about how sports is changing due to AI. What will you learn from this interview? 1. How AI Is Transforming Sports Using computer vision to analyze every movement, event, and play in real-time. Moving beyond basic stats to understanding impact and intent on the field. 2. What Makes SCORE Different Built on decentralized AI (Bittensor) and collective intelligence. Designed to work even with low-quality video—enabling access for high schools and amateur clubs. 3. Real Use Cases Player tracking, formation analysis, injury prediction, and in-game decision support. Visual tools like heat maps and frame-by-frame breakdowns. 4. Applications Beyond Pro Teams Empowering grassroots teams and scouts with elite-level insights. Parents filming Sunday league games could unknowingly be training data sources. 5. Fantasy Sports Integration AI-powered projections and analysis for fantasy leagues. Build-your-own tools for fans who want a data edge. 6. Injury Risk Detection Early signals from movement patterns that correlate with higher injury potential. Long-term value for athlete health and coaching adjustments. 7. Preparing for the AR/3D Future Compatible with lightfield displays and AR glasses (think Vision Pro). Real-time stats layered over gameplay during broadcasts. 8. The Role of Betting in Driving Innovation How sportsbooks and gambling tech are quietly pushing AI in sports forward. Inside view on how that funding and data are transforming scouting and coaching. 9. Startup Insights Bootstrapping vs. raising capital in deep tech. Hiring elite AI talent without spending $10M+ like Meta—thanks to open systems like Bittensor. 10. The Bigger Vision Creating a universal scoring system for athletes—objective, data-rich, and fair. Challenging legacy scouting reports with measurable intelligence.
23:46

AI is changing sports. Here is how. I sit down with Max Sebti, , founder and CEO of Score, and he gives me the latest about how sports is changing due to AI. What will you learn from this interview? 1. How AI Is Transforming Sports Using computer vision to analyze every movement, event, and play in real-time. Moving beyond basic stats to understanding impact and intent on the field. 2. What Makes SCORE Different Built on decentralized AI (Bittensor) and collective intelligence. Designed to work even with low-quality video—enabling access for high schools and amateur clubs. 3. Real Use Cases Player tracking, formation analysis, injury prediction, and in-game decision support. Visual tools like heat maps and frame-by-frame breakdowns. 4. Applications Beyond Pro Teams Empowering grassroots teams and scouts with elite-level insights. Parents filming Sunday league games could unknowingly be training data sources. 5. Fantasy Sports Integration AI-powered projections and analysis for fantasy leagues. Build-your-own tools for fans who want a data edge. 6. Injury Risk Detection Early signals from movement patterns that correlate with higher injury potential. Long-term value for athlete health and coaching adjustments. 7. Preparing for the AR/3D Future Compatible with lightfield displays and AR glasses (think Vision Pro). Real-time stats layered over gameplay during broadcasts. 8. The Role of Betting in Driving Innovation How sportsbooks and gambling tech are quietly pushing AI in sports forward. Inside view on how that funding and data are transforming scouting and coaching. 9. Startup Insights Bootstrapping vs. raising capital in deep tech. Hiring elite AI talent without spending $10M+ like Meta—thanks to open systems like Bittensor. 10. The Bigger Vision Creating a universal scoring system for athletes—objective, data-rich, and fair. Challenging legacy scouting reports with measurable intelligence.

Robert Scoble

65,908 просмотров • 1 год назад

Mistral just released Voxtral TTS, an open-weight voice model that tries to match ElevenLabs while keeping it free. A 3-billionn param model that fits on a laptop and runs 6X faster than real-time speech. It is split into 3 parts that handle different jobs. - The 3.4B part is the main language brain that understands the text and plans what should be spoken. - The 390M part turns that plan into speech features like rhythm, pronunciation, and tone. - The 300M part converts those speech features into the final compressed audio signal you can play. That split helps because each part only does the job it is best at, which can make the whole system faster and lighter than one giant model doing everything alone. Mistral claims about 90ms to first audio, about 6x faster than real-time speech output, about 3GB RAM after quantization, 9 languages, and voice copying from 5s of audio, even across languages. Its internal tests report 62.8% preference over ElevenLabs Flash on default voices and 69.9% on custom voices, but the bigger point is that companies can keep speech data on their own machines. IMO, enterprise buyers usually care most about control, cost, and latency, So this model should give it great push for voice-based AI apps/products.
1:21

Mistral just released Voxtral TTS, an open-weight voice model that tries to match ElevenLabs while keeping it free. A 3-billionn param model that fits on a laptop and runs 6X faster than real-time speech. It is split into 3 parts that handle different jobs. - The 3.4B part is the main language brain that understands the text and plans what should be spoken. - The 390M part turns that plan into speech features like rhythm, pronunciation, and tone. - The 300M part converts those speech features into the final compressed audio signal you can play. That split helps because each part only does the job it is best at, which can make the whole system faster and lighter than one giant model doing everything alone. Mistral claims about 90ms to first audio, about 6x faster than real-time speech output, about 3GB RAM after quantization, 9 languages, and voice copying from 5s of audio, even across languages. Its internal tests report 62.8% preference over ElevenLabs Flash on default voices and 69.9% on custom voices, but the bigger point is that companies can keep speech data on their own machines. IMO, enterprise buyers usually care most about control, cost, and latency, So this model should give it great push for voice-based AI apps/products.

Rohan Paul

15,672 просмотров • 5 месяцев назад