The world of AI is abuzz with huge developments. OpenAI has made ChatGPT’s developer mode available. Now, it can actually perform things. It means ChatGPT does more than just responding to questions. It can perform actual actions. ByteDance is also causing a stir. Its new image model is quicker and more affordable. It’s a direct…
Google released EmbeddingGemma. This new AI model redefines small AI. It performs like much bigger AI programs. Moreover, it runs completely offline on phones or laptops. It is very fast, even on basic devices. It knows over 100 languages. Moreover, it tops tests for models under 500 million parts. It works with common AI tools…
Within a week China has released two massive AI models. The Qwen team of Alibaba published a preview that performs better than the leading models. New start-up Moonshot AI is deploying its own AI with massive upgrades. The worldwide competition in AI has just become serious. We will dissect what this is. These emerging models…
Artificial Intelligence is a rapidly changing world. Something amazing was demonstrated by Microsoft. Their math AI is built much better than bigger models. Even it did not take long to train. And they have two new AI models in hand. One makes voices that are super real and the other is a language model that…
Artificial intelligence is a rapidly evolving world. Apple has just demonstrated Fast VLM. It is a fresh vision language model. It’s 85 times faster. It’s also three times smaller. And the best thing is that it is compatible with a MacBook Pro. This may be the giant leap that will allow AI to really see…
Liquid AI recently introduced LFM2-VL and it might alter the process with which we utilize AI. The new technology would enable the vision and language AI to be deployed on your smartphone, laptop, or even smartwatch. It is quick, perhaps two times quicker than other models. That is to say that AI can also be…
Introduction: The Unannounced Arrival of a Frontier Model DeepSeek V3.1 landed on Hugging Face with no fanfare and the entire world of AI was shaken instantly. This open-source model has 685 billion parameters and a context window of 128,000 tokens. What really wowed people, however, was its benchmark score, it scored higher than Claude Opus…
The AI world is abuzz, but it is not OpenAI or Google who made the huge news this time. A startup in Singapore has just released a new AI agent named HRM and it may be the largest step forward with AI reasoning since years ago. It is no larger, and not trained on more…
The beginning of mini nerves is shaking the world of large language models (LLMs) because of MiniAX M1. It was modeled to make it free, capable, and user-friendly, and does so in a way not achievable in an open-source AI. This model with its huge 1 million token context window, no usage restrictions, and more…
Artificial Intelligence is changing fast. The big boys namely OpenAI and Anthropic take the first line, but the small scale startups now make some noise as well. One of them, a startup for Europe, called Mistral, have only recently introduced a new model called Medium 3. It’s not just an incremental change, it is a…
