Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Traditional physics-based models struggle with extreme rainfall, often overestimating light events while underestimating extreme precipitation. Our new hybrid climate model, NeuralGCM, more accurately simulates extreme rainfall, mean precipitation (with a 40% bias reduction compared to CMIP6 models), and the diurnal cycle of precipitation. These improvements are a direct result... show more

Google Research

93,837 subscribers

16,267 views • 7 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Cyclone Ditwah update from Google Deepmind AI model ------------------------ Since there is lot of interest on the entry of AI models lets see what the latest AI model of Google says on the projected track of Ditwah. Note Chennai and KTCC rains will be only from late 29th to 30th November. No extreme rainfall is expected from this cyclone for Chennai. There will be heavy rainfall and a chance for very heavy rainfall too in KTCC / Chennai. Note: Please see the classification of Rainfall:- Category 24 hour rainfall over a station ending at 0830 hours ----------------------- Very Light Rain: Trace – 2.4 mm Light Rain: 2.5 – 15.5 mm (upto 2 cm) Moderate Rain: 15.6 – 64.4 mm (2 to 6 cm) Heavy Rain: 64.5 – 115.5 mm (7 to 12 cm) Very Heavy Rain: 115.6 – 204.4 mm (12 to 20 cm) Extremely Heavy Rain: ≥ 204.5 mm (> 20 cm)

Cyclone Ditwah update from Google Deepmind AI model ------------------------ Since there is lot of interest on the entry of AI models lets see what the latest AI model of Google says on the projected track of Ditwah. Note Chennai and KTCC rains will be only from late 29th to 30th November. No extreme rainfall is expected from this cyclone for Chennai. There will be heavy rainfall and a chance for very heavy rainfall too in KTCC / Chennai. Note: Please see the classification of Rainfall:- Category 24 hour rainfall over a station ending at 0830 hours ----------------------- Very Light Rain: Trace – 2.4 mm Light Rain: 2.5 – 15.5 mm (upto 2 cm) Moderate Rain: 15.6 – 64.4 mm (2 to 6 cm) Heavy Rain: 64.5 – 115.5 mm (7 to 12 cm) Very Heavy Rain: 115.6 – 204.4 mm (12 to 20 cm) Extremely Heavy Rain: ≥ 204.5 mm (> 20 cm)

Tamil Nadu Weatherman

46,058 views • 10 months ago

We are releasing the fastest TTS model as open source! One of the biggest problems we encountered while optimizing TTS models is optimization itself. Since it is LLM-based, even optimizing with VLLM and SgLang libraries is not enough. Additionally, I made the outputs much better with special optimizations for training. The voices you are listening to are results obtained from only 20% of the model training. We trained these outputs on 8xB200 in 24 hours. Even though the dataset includes some low-quality data, the model's output is much better. We will release the code as open source tomorrow. Data: Emilia-3m Model: EchoDit (custom) Codec: DacVae Opt: Flash-attn + CacheDit

We are releasing the fastest TTS model as open source! One of the biggest problems we encountered while optimizing TTS models is optimization itself. Since it is LLM-based, even optimizing with VLLM and SgLang libraries is not enough. Additionally, I made the outputs much better with special optimizations for training. The voices you are listening to are results obtained from only 20% of the model training. We trained these outputs on 8xB200 in 24 hours. Even though the dataset includes some low-quality data, the model's output is much better. We will release the code as open source tomorrow. Data: Emilia-3m Model: EchoDit (custom) Codec: DacVae Opt: Flash-attn + CacheDit

Kadir Nar

32,232 views • 2 months ago

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

Sundar Pichai

864,841 views • 1 year ago

Space objects in 3D NASA created 3D models of stars and supernova remnants by combining data from the Chandra X-ray Observatory with computer calculations. These models can not only be viewed on the screen, but also printed on a 3D printer. Here they are, from left to right — the supernova remnant Cassiopeia A, the young star BP Tauri, the planetary nebula Cygnus Loop, and the supernova remnant G292.0+1.8.

Space objects in 3D NASA created 3D models of stars and supernova remnants by combining data from the Chandra X-ray Observatory with computer calculations. These models can not only be viewed on the screen, but also printed on a 3D printer. Here they are, from left to right — the supernova remnant Cassiopeia A, the young star BP Tauri, the planetary nebula Cygnus Loop, and the supernova remnant G292.0+1.8.

Black Hole

13,099 views • 1 year ago

The #Brabus Bodo is directly related to Aston Martin. It is a coachbuilt hyper-GT created by #Brabus, based on the #AstonMartin Vanquish (the current V12 model). #Brabus uses the Vanquish’s platform, chassis, powertrain, and many interior components, then completely reworks it with a new full-carbon-fiber body, extreme tuning, and custom design to produce around 1,000 hp (986 bhp), 1,200 Nm of torque, and a top speed of approximately 360 km/h.

The #Brabus Bodo is directly related to Aston Martin. It is a coachbuilt hyper-GT created by #Brabus, based on the #AstonMartin Vanquish (the current V12 model). #Brabus uses the Vanquish’s platform, chassis, powertrain, and many interior components, then completely reworks it with a new full-carbon-fiber body, extreme tuning, and custom design to produce around 1,000 hp (986 bhp), 1,200 Nm of torque, and a top speed of approximately 360 km/h.

HyperCar™

77,418 views • 4 months ago

A major question in multimodal modeling is how to leverage strong pre-trained models, such as industry-scale LLMs and VLMs, during training to avoid starting from scratch. This is important as the demand for rich and domain-specific multimodal models continues to increase, while training them from scratch is obviously impractical due to limited data and compute. The so-called “any-to-any” multimodal models can model a large diverse dictionary of modalities. That’s good. Their downside is that their architectures are often not decoder-only, which has limited their performance in practice and prevents them from leveraging strong pre-trained decoder-only models as priors. We are releasing MODUS (ICML Conference '26) a decoder-only any-to-any multimodal model to address some of these questions. A single transformer decoder predicts any modality from any others with no modality-specific heads, losses, or task pipelines. We show efficient adaptation of established models, e.g., BAGEL, to rich any-to-any multimodal modeling. We are releasing 14B to 77B parameter models. All materials are open-source. The download links and demos here 🧵

A major question in multimodal modeling is how to leverage strong pre-trained models, such as industry-scale LLMs and VLMs, during training to avoid starting from scratch. This is important as the demand for rich and domain-specific multimodal models continues to increase, while training them from scratch is obviously impractical due to limited data and compute. The so-called “any-to-any” multimodal models can model a large diverse dictionary of modalities. That’s good. Their downside is that their architectures are often not decoder-only, which has limited their performance in practice and prevents them from leveraging strong pre-trained decoder-only models as priors. We are releasing MODUS (ICML Conference '26) a decoder-only any-to-any multimodal model to address some of these questions. A single transformer decoder predicts any modality from any others with no modality-specific heads, losses, or task pipelines. We show efficient adaptation of established models, e.g., BAGEL, to rich any-to-any multimodal modeling. We are releasing 14B to 77B parameter models. All materials are open-source. The download links and demos here 🧵

Amir Zamir

24,161 views • 3 months ago

World modeling and imitation learning have largely been considered two disparate worlds. In our recent work, Unified World Models, just accepted to #RSS2025, Chuning Zhu provides a dead-simple unifying solution: just train a joint diffusion model over actions and future states, but with *decoupled* diffusion time steps across these modalities. Manipulating these decoupled time steps then allows for marginalization or conditioning on actions or states; a single model can serve as a policy, forward dynamics model, video prediction model, or inverse dynamics model by simply setting diffusion timesteps carefully. The resulting model can leverage video datasets along with robot training data much more effectively, and shows improved robustness, generalization, and flexibility. This is exciting because it is frustratingly simple, scalable, and shows strong improvement on real-world robotics problems. Please refer to Chuning Zhu 's excellent thread for more details! More details/code can be found on our website and in the paper -

World modeling and imitation learning have largely been considered two disparate worlds. In our recent work, Unified World Models, just accepted to #RSS2025, Chuning Zhu provides a dead-simple unifying solution: just train a joint diffusion model over actions and future states, but with decoupled diffusion time steps across these modalities. Manipulating these decoupled time steps then allows for marginalization or conditioning on actions or states; a single model can serve as a policy, forward dynamics model, video prediction model, or inverse dynamics model by simply setting diffusion timesteps carefully. The resulting model can leverage video datasets along with robot training data much more effectively, and shows improved robustness, generalization, and flexibility. This is exciting because it is frustratingly simple, scalable, and shows strong improvement on real-world robotics problems. Please refer to Chuning Zhu 's excellent thread for more details! More details/code can be found on our website and in the paper -

Abhishek Gupta

11,430 views • 1 year ago

Video diffusion models have strong implicit representations of 3D shape, material, and lighting, but controlling them with language is cumbersome, and control is critical for artists and animators. GenLit connects these implicit representations with a continuous 5D control signal describing the direction and intensity of a point light source. This enables single-image near-field relighting of an image using a video diffusion model. We use a ControlNet-like approach and show that, with a small amount of synthetic data, GenLit generalizes to complex real-world images. Given a single image and the 5D lighting signal, GenLit creates a video of a moving light source that is inside the scene. It moves around and behind scene objects, producing effects such as shading, cast shadows, secularities, and interreflections with a realism that is hard to obtain with traditional inverse rendering methods. GenLit shows that it is possible to get continuous control over implicit physical processes within a video model. I think this is just the beginning and promises to make such models much more practical for creators. Shrisha Bharadwaj will present today at SIGGRAPH Asia Room: S423/S424, Level 4 @ 13:50 on 15 of Dec.

Video diffusion models have strong implicit representations of 3D shape, material, and lighting, but controlling them with language is cumbersome, and control is critical for artists and animators. GenLit connects these implicit representations with a continuous 5D control signal describing the direction and intensity of a point light source. This enables single-image near-field relighting of an image using a video diffusion model. We use a ControlNet-like approach and show that, with a small amount of synthetic data, GenLit generalizes to complex real-world images. Given a single image and the 5D lighting signal, GenLit creates a video of a moving light source that is inside the scene. It moves around and behind scene objects, producing effects such as shading, cast shadows, secularities, and interreflections with a realism that is hard to obtain with traditional inverse rendering methods. GenLit shows that it is possible to get continuous control over implicit physical processes within a video model. I think this is just the beginning and promises to make such models much more practical for creators. Shrisha Bharadwaj will present today at SIGGRAPH Asia Room: S423/S424, Level 4 @ 13:50 on 15 of Dec.

Michael Black

22,353 views • 9 months ago

World Models are the path for some AI Models in the future. But how can we efficiently train these models to not only see the world the way humans do but to see the world in a new and unique way. By visualizing, what is normally sequenced audio patterns, we can derive much more insights. Here we see Paganini in a visual form that can than be described and transcribed into a World Model. We can observe connections in a manner that may not have been clear prior to the digitalization of music and sound in this way. The company with the most valuable potential in building a World Model is Tesla. Not that this type of visualization is being used, but that the mechanisms are in place, and the technology is in place for the company to thrive in this new form of AI.

World Models are the path for some AI Models in the future. But how can we efficiently train these models to not only see the world the way humans do but to see the world in a new and unique way. By visualizing, what is normally sequenced audio patterns, we can derive much more insights. Here we see Paganini in a visual form that can than be described and transcribed into a World Model. We can observe connections in a manner that may not have been clear prior to the digitalization of music and sound in this way. The company with the most valuable potential in building a World Model is Tesla. Not that this type of visualization is being used, but that the mechanisms are in place, and the technology is in place for the company to thrive in this new form of AI.

Brian Roemmele

57,454 views • 10 months ago

AI needs data to grow. But accessing high-quality datasets is becoming harder than ever. · Data collection is increasingly centralized in the hands of a few large players. · Platforms are locking content behind expensive APIs and private data deals. · Specialized datasets (medical, legal, code) can cost 20–40× more than general web data. · And while 94% of companies are exploring Gen-AI, only 12% have reliable data channels. The result? A growing data bottleneck for AI development. The future of AI won’t just depend on better models. It will depend on who can access and activate quality data.

AI needs data to grow. But accessing high-quality datasets is becoming harder than ever. · Data collection is increasingly centralized in the hands of a few large players. · Platforms are locking content behind expensive APIs and private data deals. · Specialized datasets (medical, legal, code) can cost 20–40× more than general web data. · And while 94% of companies are exploring Gen-AI, only 12% have reliable data channels. The result? A growing data bottleneck for AI development. The future of AI won’t just depend on better models. It will depend on who can access and activate quality data.

Perceptron Network

75,490 views • 6 months ago

✨ Every time the video models get better, the try on model on Photo AI also becomes a lot more useful, as a large % of my customers now are e-commerce store And showing clothes in a video is nice for sales! With AI this means stores don't need to do expensive shoots flying a model and entire camera and light crew around the world They can just upload a few photos of their models, then upload the clothes, and describe the setting (like a beach in Thailand) and in less than 10 seconds it's generated, for a video in less than a minute! Below is the input: a dress laid flat, and output: a full video shoot

✨ Every time the video models get better, the try on model on Photo AI also becomes a lot more useful, as a large % of my customers now are e-commerce store And showing clothes in a video is nice for sales! With AI this means stores don't need to do expensive shoots flying a model and entire camera and light crew around the world They can just upload a few photos of their models, then upload the clothes, and describe the setting (like a beach in Thailand) and in less than 10 seconds it's generated, for a video in less than a minute! Below is the input: a dress laid flat, and output: a full video shoot

@levelsio

334,317 views • 1 year ago

Building on the previous paper, in this study we compare a continuous “smooth return” S2>S1 model with an event-driven one, where long periods of relative calm are punctuated by short, intense episodes of global reorganisation. Both models cover the same time window. Neither uses archaeological data in its construction. When compared against where early humans and early civilizations actually appear and persist, the difference is statistically robust. The smooth model behaves like background noise. The event-driven model lines up in time and space far better than chance allows, even after aggressive temporal and spatial randomization tests. Statistically, the event-driven model lines up with where and when early civilizations appear far better than a smooth, continuous model, even after we randomize both timing and location to test what could arise by chance. The event timeline itself was built independently from well-known late-glacial disruptions - such as Heinrich events, meltwater pulses, and abrupt deglacial transitions - rather than from any archaeological data. Nothing here claims that specific events caused specific cultures. It does suggest that history may not unfold on a smooth clock. Human societies seem to flourish during recovery phases between disruptions, not during the disruptions themselves. The animation contrasts the two return models. Draft paper : Source & Results : (coming soon)

Building on the previous paper, in this study we compare a continuous “smooth return” S2>S1 model with an event-driven one, where long periods of relative calm are punctuated by short, intense episodes of global reorganisation. Both models cover the same time window. Neither uses archaeological data in its construction. When compared against where early humans and early civilizations actually appear and persist, the difference is statistically robust. The smooth model behaves like background noise. The event-driven model lines up in time and space far better than chance allows, even after aggressive temporal and spatial randomization tests. Statistically, the event-driven model lines up with where and when early civilizations appear far better than a smooth, continuous model, even after we randomize both timing and location to test what could arise by chance. The event timeline itself was built independently from well-known late-glacial disruptions - such as Heinrich events, meltwater pulses, and abrupt deglacial transitions - rather than from any archaeological data. Nothing here claims that specific events caused specific cultures. It does suggest that history may not unfold on a smooth clock. Human societies seem to flourish during recovery phases between disruptions, not during the disruptions themselves. The animation contrasts the two return models. Draft paper : Source & Results : (coming soon)

Craig Stone

10,899 views • 8 months ago

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

Axis Robotics

28,096 views • 9 months ago

A metal origami! 🪭 This method is called Hyperbolic Metal Forming and is hypnotizing to watch. Instead of shaping metal with slow mechanical force, HMF uses controlled shockwaves to form complex geometries at extreme speed, often without the need for heavy dies or post-processing. The result is stronger, lighter parts with shapes that are almost impossible using traditional stamping. That’s why you see it popping up in aerospace, automotive structures, and defense components. Think of it like metal origami, but driven by high-energy pulses instead of presses. A small reminder that some things in manufacturing come from physics, not just automation. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

A metal origami! 🪭 This method is called Hyperbolic Metal Forming and is hypnotizing to watch. Instead of shaping metal with slow mechanical force, HMF uses controlled shockwaves to form complex geometries at extreme speed, often without the need for heavy dies or post-processing. The result is stronger, lighter parts with shapes that are almost impossible using traditional stamping. That’s why you see it popping up in aerospace, automotive structures, and defense components. Think of it like metal origami, but driven by high-energy pulses instead of presses. A small reminder that some things in manufacturing come from physics, not just automation. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

567,473 views • 10 months ago

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

Depth Any Video with Scalable Synthetic Data AI physicists and chemists continue to make strides in depth estimation from video. Check out this new paper featuring some impressive examples. See the thread for more details (unfortunately no code yet). Abstract: Video depth estimation has long been hindered by the scarcity of consistent and scalable ground truth data, leading to inconsistent and unreliable results. In this paper, we introduce Depth Any Video, a model that tackles the challenge through two key innovations. First, we develop a scalable synthetic data pipeline, capturing real-time video depth data from diverse game environments, yielding 40,000 video clips of 5-second duration, each with precise depth annotations. Second, we leverage the powerful priors of generative video diffusion models to handle real-world videos effectively, integrating advanced techniques such as rotary position encoding and flow matching to further enhance flexibility and efficiency. Unlike previous models, which are limited to fixed-length video sequences, our approach introduces a novel mixed-duration training strategy that handles videos of varying lengths and performs robustly across different frame rates 0 - even on single frames. At inference, we propose a depth interpolation method that enables our model to infer high-resolution video depth across sequences of up to 150 frames. Our model outperforms all previous generative depth models in terms of spatial accuracy and temporal consistency.

MrNeRF

27,428 views • 2 years ago

Generative 3d environments just became a thing with the announcement of OpenAI's new video model, Sora. Michael Rublof from took one of those videos, and turned it into a NeRF using Colmap and Nerfstudio. While people are laughing at the topology of generated models, the world is changing around us, and that's exciting and a little scary, but we'll find a way to turn Gen Ai into creative superpowers. I believe in human creativity, in our ability to surprise, to move, to change and to challenge. Here's to the future! #ai #art

Generative 3d environments just became a thing with the announcement of OpenAI's new video model, Sora. Michael Rublof from took one of those videos, and turned it into a NeRF using Colmap and Nerfstudio. While people are laughing at the topology of generated models, the world is changing around us, and that's exciting and a little scary, but we'll find a way to turn Gen Ai into creative superpowers. I believe in human creativity, in our ability to surprise, to move, to change and to challenge. Here's to the future! #ai #art

Martin Nebelong

288,988 views • 2 years ago

How to make money on the weather using Polymarket I've been noticing more and more traders quietly printing on Polymarket's weather markets lately - and the category is exploding for a reason. Weather has always been super predictable for meteorologists (and us normals) right up to the day of. The whole point? You can earn easy, near-certain yield just knowing it'll rain in London tomorrow. I'm sharing a finished tutorial with you. Here is a small list of traders 1. gopfan2 ( - The absolute leader in weather. Earned over $2M in net profit by focusing on temperature and precipitation. Strategy - buy Yes below 15 cents, No above 45 cents, with risks of less than $1 per position. It dominates the NYC and London markets where the weather is predictable 2. enzocostapt81 ( is a weather-exclusive trader whose profile shows a complete wipeout in resolved positions-no active/open trades, current positions value $0.00, and all listed markets (resolved) at -100% P/L. The trader focused solely on daily/precise temperature predictions in major cities like New York City and London. 3. 0x594edb9112f526fa6a80b8f858a6379c8a2c1c11 ( 100% of active positions are weather/temperature markets across cities like Dallas, London, Seattle, Atlanta, NYC, and Toronto—focused on precise daily highs/thresholds/ranges. 4. meropi ( - Earned ~$30k on micro bets ($1-3) with multipliers up to 500x. Automated bets on temperature rise for 0.01 cents. Focus on speed to capture momentum in daily markets. One of the most stable in weather 5. 1pixel ( – $18.5k profit from $2.3k deposit, weather only (NYC and London) 6. erb80 ( Dominant focus-two massive Atlanta temperature range bets for Dec 17, with enormous share volume at ultra-low entries (0.1¢) turning into huge unrealized gains (+49,550% on the main one) 7. Hans323 ( - Earned $1.1M on one temperature trade in London. Started with $741 in January 2025 and increased to $87k net profit for the year 8. securebet ( - Turned $7 into $640 (+9244%) on a series of temperature bets in NYC and Seattle. 3077 predictions, top 0.04% by metrics. Focus on small bets ($3-20) with high growth on low quotes. High win rate thanks to NOAA data 9. automatedAItradingbot ( Micro/low-cost bets (0.4¢–15¢) on specific outcomes, especially weather thresholds in Seoul/London and fighter matchups.Explosive wins (300%+ on select weather 1,000–5,000% average ROI across successful weather specialists based on this traders Tools and Automation > ( - Built specifically for Polymarket weather traders. Offers real-time multi-model forecasts (GFS, ECMWF, etc.), temperature range dashboards, climate pattern guides per city/station, and settlement station details. Includes educational guides on seasonal biases and forecasting challenges—highly recommended for NYC/London/Atlanta markets. > ( — Free guide/resource hub for weather betting on Polymarket. Covers market overviews, settlement rules >Tropical Tidbits ( - US GFS and ECMWF Europe models for temperature, precipitation, hurricane forecasts. Updates every 6 hours. Ideal for comparing models if 3+ agree, the probability is high >Climate Reanalyzer ( - real-time maps of air/ocean temperature, precipitation anomalies. With historical context for calculating probabilities >Windy ( - interactive maps of wind, temperature, rain, snow. 10+ models, for local events NOAA Climate Data Online ( - 100+ years of historical location data NOAA Weather Prediction ? >Center ( - short forecasts for precipitation, anomalies. Climate Prediction Center ( - long-term ENSO, droughts >Open-Meteo ( - Completely free open-source weather API with no key required. Provides GFS, ECMWF-derived, and ensemble forecasts for temperature, precipitation, and more at hourly/sub-hourly resolution globally. Excellent for scripting quick checks on NYC/London highs or comparing multiple models. Direct API calls make it ideal for automation or batch probability calculations. >OpenWeatherMap ( = Free tier gives current conditions, 5-day/3-hour forecasts, and 16-day daily forecasts. Good for real-time verification and basic historical pulls (limited free). Use for cross-checking Polymarket ranges before resolution. >Visual Crossing Weather ( - Free tier includes historical data (50+ years), current conditions, hourly/sub-hourly forecasts, and alerts. Strong for querying specific cities >WeatherAPI. com ( - Free plan covers real-time, hourly, daily forecasts (up to 14 days), historical data (from 2010), and bulk requests. Reliable for urban stations and includes marine/pollen extras if needed. Quick Tips for Using These in Trading >>>Cross-verify 3+ models (e.g., GFS + ECMWF via Open-Meteo + Windy) → if 80%+ agree on a range/threshold, probability is often very high for "Yes" bets under 10-15¢. >>>Focus on major stations (e.g., Central Park for NYC, Heathrow for London) - check settlement rules on Polymarket pages. >>>ADD TO BOOKMARKS so you don't lose alpha information

Aleiah

77,547 views • 8 months ago

It is getting cold, blustery and wet out there! 📡🥶 If you were watching the National Weather Service (NWS) radar data out of NWS Tampa Bay this afternoon, you would have noticed what appeared to be a sharp increase in the coverage of showers from one frame to the next over the Tampa Bay region. 📡 ☔️ In reality, the precipitation was not increasing. The trained NWS meteorologist and radar operator shifted the radar scanning strategy to better detect lighter, lower-level precipitation. Notice the change at the top of this radar animation with the “VCP”, or volume coverage pattern. Watch it change from VCP 215 to VCP 34. What does this mean? 🤔 The National Weather Service Doppler radar volume coverage pattern (VCP) modes determine how radar beams are emitted and the angle at which they scan the atmosphere. VCP Mode 215: This mode provides higher vertical resolution due to more elevation angles being processed. It typically includes more volume scans, focusing on detecting severe weather phenomena such as thunderstorms and tornadoes effectively. VCP Mode 34: This mode offers lower vertical resolution with fewer elevation angles. It is designed for less demanding weather situations, providing broader coverage and is often used during more stable weather conditions. As these sprinkles and light showers continue to develop and move ashore from the Gulf within a blustery northwest wind, much colder (and drier) air will overspread the area through tonight into Sunday morning when the potential exists for isolated flurries or frozen rain drops (i.e., sleet or ice pellets). VCP 34 typically features more lower elevation scanning angles compared to VCP 215. 📡Things to remember📡: Lower radar elevation angles sampled by VCP 34 are designed to scan the atmosphere closer to the ground. This design is beneficial for detecting broader weather patterns and identifying precipitation types that occur near the surface. Stay warm! #FLwx #Gasparilla City of Tampa Hillsborough County

It is getting cold, blustery and wet out there! 📡🥶 If you were watching the National Weather Service (NWS) radar data out of NWS Tampa Bay this afternoon, you would have noticed what appeared to be a sharp increase in the coverage of showers from one frame to the next over the Tampa Bay region. 📡 ☔️ In reality, the precipitation was not increasing. The trained NWS meteorologist and radar operator shifted the radar scanning strategy to better detect lighter, lower-level precipitation. Notice the change at the top of this radar animation with the “VCP”, or volume coverage pattern. Watch it change from VCP 215 to VCP 34. What does this mean? 🤔 The National Weather Service Doppler radar volume coverage pattern (VCP) modes determine how radar beams are emitted and the angle at which they scan the atmosphere. VCP Mode 215: This mode provides higher vertical resolution due to more elevation angles being processed. It typically includes more volume scans, focusing on detecting severe weather phenomena such as thunderstorms and tornadoes effectively. VCP Mode 34: This mode offers lower vertical resolution with fewer elevation angles. It is designed for less demanding weather situations, providing broader coverage and is often used during more stable weather conditions. As these sprinkles and light showers continue to develop and move ashore from the Gulf within a blustery northwest wind, much colder (and drier) air will overspread the area through tonight into Sunday morning when the potential exists for isolated flurries or frozen rain drops (i.e., sleet or ice pellets). VCP 34 typically features more lower elevation scanning angles compared to VCP 215. 📡Things to remember📡: Lower radar elevation angles sampled by VCP 34 are designed to scan the atmosphere closer to the ground. This design is beneficial for detecting broader weather patterns and identifying precipitation types that occur near the surface. Stay warm! #FLwx #Gasparilla City of Tampa Hillsborough County

Brian LaMarre

12,298 views • 8 months ago

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

Wonderland: Navigating 3D Scenes from a Single Image Contributions: • First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. • Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. • Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasks—and bridging image space and 3D space—through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256× spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

MrNeRF

52,849 views • 1 year ago