Nvidia released Nemotron-Ultra, a 253B parameter reasoning AI —Surpasses DeepSeek R1, Llama 4 Behemoth, and Maverick across benchmarks —Includes a reasoning on/off toggle
—Open-source with model code, weights, and post-training data on Hugging Face
@rowancheung
-
Nvidia Nemotron-Ultra 253B Parameter Reasoning AI Surpasses Competitors
By
–
-
Google Launches Deep Research on Gemini 2.5 Pro
By
–
Google made headlines by making Deep Research available on Gemini 2.5 Pro Exp The move enables Gemini to create superior research reports over rivals Also includes new audio overviews to turn reports into podcast-like conversations!
-
Amazon Upgrades Nova Reel 1.1 Video Model with Extended Capabilities
By
–
Amazon also dropped an upgraded Nova Reel 1.1 video model
— Rowan Cheung (@rowancheung) 9 avril 2025
—Delivers improved quality, style consistency
—Extends generations to 2 min via automated and manual, shot-by-shot modes
—Also available on Amazon Bedrockpic.twitter.com/eiNfOzLlLaAmazon also dropped an upgraded Nova Reel 1.1 video model —Delivers improved quality, style consistency
—Extends generations to 2 min via automated and manual, shot-by-shot modes
—Also available on Amazon Bedrock -
Amazon Nova Sonic Speech-to-Speech AI Outperforms OpenAI Models
By
–
Amazon launched Nova Sonic speech-to-speech AI for human-like interactions
— Rowan Cheung (@rowancheung) 9 avril 2025
—Outperforms OpenAI's voice models with ~ 80% less cost
—4.2% word error rate across languages
— 46.7% better accuracy than GPT-4o for noisy environments
—On Amazon Bedrockpic.twitter.com/5e5nAeUXhvAmazon launched Nova Sonic speech-to-speech AI for human-like interactions —Outperforms OpenAI's voice models with ~ 80% less cost
—4.2% word error rate across languages
— 46.7% better accuracy than GPT-4o for noisy environments
—On Amazon Bedrock -
Amazon Releases New Voice Model Surpassing OpenAI
By
–
TODAY'S AI NEWS: Amazon just dropped a new voice model that beats OpenAI Plus, more news from Google, Nvidia, Deep Cogito, Stanford, and more. Here's everything you need to know:
-
MedSAM2: AI Model for 3D Medical Image Segmentation
By
–
Harvard and University of Toronto researchers dropped an AI for 3D medical image & video segmentation
— Rowan Cheung (@rowancheung) 8 avril 2025
Built atop Segment Anything Model 2.1, MedSAM2 generalizes across organs, modalities, and pathologies with an 85%+ reduction in annotation costspic.twitter.com/ZW66mTldfjHarvard and University of Toronto researchers dropped an AI for 3D medical image & video segmentation Built atop Segment Anything Model 2.1, MedSAM2 generalizes across organs, modalities, and pathologies with an 85%+ reduction in annotation costs
-
Krea AI Raises $83M for AI Image and Video Platform
By
–
Krea AI just raised $83M in funding from Bain Capital Ventures and others
— Rowan Cheung (@rowancheung) 8 avril 2025
The startup is building a unified browser-based platform that brings together tools to generate, edit, and customize AI-generated images and videospic.twitter.com/u4wQFABN8lKrea AI just raised $83M in funding from Bain Capital Ventures and others The startup is building a unified browser-based platform that brings together tools to generate, edit, and customize AI-generated images and videos
-
UC Berkeley Stanford researchers use Test Time Training for video generation
By
–
Researchers from UC Berkeley and Stanford tapped Test Time Training to generate one-minute videos
— Rowan Cheung (@rowancheung) 8 avril 2025
They added TTT layers to a pre-trained Transformer and fine-tuned it to generate cartoons with strong temporal consistency
Here's a single-shot example:pic.twitter.com/YCqNTHSvpuResearchers from UC Berkeley and Stanford tapped Test Time Training to generate one-minute videos They added TTT layers to a pre-trained Transformer and fine-tuned it to generate cartoons with strong temporal consistency Here's a single-shot example:
-
ElevenLabs MCP Server Enables AI Voice Agents via Claude
By
–
ElevenLabs introduced its official MCP server
— Rowan Cheung (@rowancheung) 8 avril 2025
The integration enables platforms like Claude and Cursor to access AI voice capabilities via simple text prompts
It can even be used to create automated agents for tasks like performing outbound calls!pic.twitter.com/yJMlFfUcP6ElevenLabs introduced its official MCP server The integration enables platforms like Claude and Cursor to access AI voice capabilities via simple text prompts It can even be used to create automated agents for tasks like performing outbound calls!
-
Runway Gen-4 Turbo Produces 10-Second Videos in 30 Seconds
By
–
Runway released Gen-4 Turbo, a faster version of its new AI video model
— Rowan Cheung (@rowancheung) 8 avril 2025
It can produce 10-second videos in just 30 seconds.
Now rolling out across all plans, including the free 'Basic' tier with 125 one-time credits.pic.twitter.com/ju4B7v8yhkRunway released Gen-4 Turbo, a faster version of its new AI video model It can produce 10-second videos in just 30 seconds. Now rolling out across all plans, including the free 'Basic' tier with 125 one-time credits.