Meta has launched Llama 4, a new series of AI models, on April 5, 2025, featuring Scout, Maverick, Behemoth, and L4emoth. These models focus on multimodal capabilities and efficiency via a mixture-of-experts (MoE) architecture, which allows them to perform tasks more effectively while cutting costs. Scout has 109 billion parameters and excels in processing lengthy ...
Meta has launched its latest AI models, Llama 4, on April 5, 2025. The new series includes four models: Scout, Maverick, Behemoth, and L4emoth, which emphasize multimodal capabilities and efficiency using a mixture-of-experts architecture. Scout and Maverick are available now, with Scout handling lengthy documents and Maverick focusing on creative writing and multilingual tasks. Behemoth ...
In honor of Microsoft’s 50th anniversary, GitHub is launching Agent Mode in Visual Studio Code, now available to all users with MCP support. This upgrade introduces the new GitHub Copilot Pro+ plan, featuring premium requests and access to advanced models from Anthropic, Google, and OpenAI. Users can now enjoy next edit suggestions for code completions ...
Anthropic has teamed up with Databricks in a new five-year partnership to help businesses integrate generative AI into their operations. This collaboration will offer Claude AI models through the Databricks Data Intelligence Platform, making it accessible to over 10,000 organizations. The latest Claude 3.7 Sonnet model features advanced reasoning capabilities, enabling it to provide detailed ...
Databricks has formed a five-year strategic partnership with Anthropic AI to enhance their Data Intelligence Platform by offering Anthropic’s latest AI models, including Claude 3.7 Sonnet. This collaboration allows businesses to utilize advanced AI reasoning tools through major cloud providers like Amazon Web Services, Azure, and Google Cloud. The partnership aims to help enterprises develop, ...
At the GTC 2025 event, Nvidia CEO Jensen Huang highlighted the urgent need for 100 times more computational power to support emerging reasoning models like DeepSeek-R1. He introduced advanced AI computing platforms, including the GB300 NVL72, which features the new Blackwell Ultra GPU, set to launch in late 2025. Innovations also include air-cooled data center ...
Nvidia has launched a new series of AI models called Llama Nemotron, aimed at helping businesses develop AI agents. These models, which build on Meta’s Llama language models, are enhanced for complex tasks like math and coding. Available in three sizes—Nano, Super, and Ultra—these models cater to different deployment needs. Major companies like Microsoft, SAP, ...
Nvidia Corp. is advancing in the world of agentic artificial intelligence by unveiling its new Llama Nemotron AI models at the GTC 2025 conference. These models, enhanced from Meta’s Llama models, feature improved reasoning abilities, achieving 20% greater accuracy and five times the inference speed. Available through Nvidia’s NIM microservices in three sizes—Nano, Super, and ...
Baidu, the Chinese tech giant, has unveiled two new AI models: Ernie X1 and Ernie 4.5. The company claims these models compete with offerings from OpenAI and DeepSeek while being more cost-effective. Ernie X1, a reasoning model, is said to deliver performance comparable to DeepSeek R1 at half the price, while Ernie 4.5 outperforms GPT-4.5 ...
Baidu, the leading Chinese tech company, has launched two new AI models called Ernie X1 and Ernie 4.5. These models offer competitive performance at lower prices compared to OpenAI and DeepSeek’s offerings. Ernie X1 is touted as delivering similar reasoning capabilities to DeepSeek R1 at half the cost, while Ernie 4.5 reportedly outperforms OpenAI’s GPT-4.5 ...
