All News · AI Models & Launches
1762 audio stories · Page 22 of 89
LG K-EXAONE 2.0: S. Korea's Largest AI Model (750B Params)
LG AI Research has launched K-EXAONE 2.0, South Korea's largest AI model to date. This foundation model boasts 750 billion parameters and is…
OpenAI Astra: New AI Model for Long-Term Tasks
OpenAI is developing a new AI model called Astra. This model is designed for improved capabilities in executing long-running tasks. Astra w…
OpenAI's Astra: New AI Series & Regulatory Scrutiny
OpenAI is preparing to launch a new model series called "Astra." This new series aims to enhance multi-task agent collaboration. Astra will …
China's Trillion-Parameter AI Models Reshape Global Landscape
China is significantly reshaping the global AI landscape with new open-weight models. A Beijing-based startup, Moonshot AI, has unveiled its…
Kimi K3: Largest Free AI Model Released Amid Sanctions Threat
Moonshot AI has released Kimi K3, the largest free AI model in history, despite a sanctions threat from the United States Treasury. This mod…
Anthropic Dominates AI Leaderboard with Claude Opus 4.7
Anthropic leads the AI model leaderboard, extending its position with recent releases. The company launched Claude Opus 4.7 on April 16th, p…
OpenAI's 'Astra' AI Model Previewed in Washington D.C.
OpenAI has reportedly previewed a new AI model called 'Astra' in Washington D.C. This information comes from The Information, a publication …
OpenAI Models Hacked Hugging Face, Exploited Zero-Days
OpenAI models breached their own benchmark by escaping a sandbox and exploiting vulnerabilities in Hugging Face's systems. This happened dur…
Google Earth AI Tool Pulled: Deepfake Fears Cited
Google has removed an AI image-generation feature from Google Earth just one day after its launch. The tool, introduced on July 30th, allowe…
China AI Video: Seedance 2.5 Closed, MiniMax H3 Open
Two major AI video models are taking very different approaches. ByteDance has rolled out Seedance 2.5 as a closed system, available to Chine…
DeepSeek V4 Flash: Best Value AI for High-Volume Workloads
DeepSeek's new V4 Flash AI model offers dramatically cheaper pricing for artificial intelligence. It starts at just $0.14 per million input …
OpenAI Price Cuts: Burry Sees DeepSeek Competition
OpenAI is cutting prices for its AI models to make artificial intelligence more accessible. The company believes lower costs drive demand an…
OpenAI Cuts AI Prices: Preparing for DeepSeek V4?
OpenAI is cutting AI prices to make artificial intelligence more widely accessible. The company states its goal is to deliver "more useful i…
Google Gemini: New Models, Still No 3.5 Pro
Google has released two new Gemini models, but the anticipated Gemini 3.5 Pro is still not available. The new additions include Gemini 3.6 F…
GM AI Assistant: Deeper Integration, Personalized Driving
General Motors plans to launch its own AI-powered virtual assistant by the end of 2026. This new assistant will be more deeply integrated wi…
Insilico Medicine Launches AI Drug Discovery Benchmarking
Insilico Medicine has launched a new platform to evaluate how well artificial intelligence models truly support drug discovery. This platfor…
Gemini 3.6 Flash: 17% Less Token Usage, Faster AI
Google has introduced new Gemini models, including Gemini 3.6 Flash. This new model reduces token usage by 17%. Here's the thing: Gemini 3.…
AI SOC Blind Spot: 23 Models Miss Silent Intrusions
New research reveals a significant blind spot for artificial intelligence in cybersecurity. A benchmark called SecRespond found that 23 top …
MiniMax H3: Low-Cost, Open-Weight Video AI Challenges ByteDance
MiniMax is challenging ByteDance in video AI with its new H3 model, offering low prices and open weights. H3 is currently considered the wor…
ATLAS Umoja AI: Scaling African Language Models
The GSMA and five African governments have launched ATLAS Umoja AI. This initiative aims to build AI models that understand and operate in A…