Hub AI in practice
Artificial Intelligence is present in everyday life – from “googling” to facial recognition to vacuum cleaner robots. AI tools are becoming more and more elaborate and support people and companies more effectively in their tasks, such as generating graphics, texting or coding, or interpreting large amounts of data.
What AI tools are there, how do they work, how do they help in our everyday world – and how do they change our lives? These are the questions we address in our Content Hub Artificial Intelligence in Practice.
Google's AI infrastructure is under strain as demand for its latest models increases. Product manager Logan Kilpatrick responded to complaints about the limited availability of Gemini 2.5 Pro Deep Think, explaining, "the release is constrained because this is a big model and takes a boat load of compute to run, when our TPU's are already burning to keep up with massive growth on Veo, Gemini 2.5 pro, AI mode rollout to hundreds of millions, etc."
Kilpatrick addressed criticism after users pointed out that, despite strong benchmark scores, Gemini 2.5 Pro Deep Think is difficult to use to use due to access restrictions. Even Ultra subscribers can only make a handful of requests per day as the system struggles to keep up with demand.
Wan2.2 A14B now tops the rankings for open source video models, according to Artificial Analysis. It ranks seventh for text-to-video and fourteenth for image-to-video, with the lower placement in the latter likely due to its 16 frames per second output compared to 24 fps in some competitors. Among open models, Wan2.2 A14B leads the field, but it still trails behind closed models like Veo 3 and Seedance 1.0 in overall performance. Pricing, however, is often much lower depending on the provider.
Uber Eats now manipulates food images using generative AI.
Uber Eats is now using generative AI to identify and enhance low-quality food photos on its menus. The technology does more than just adjust lighting, resolution, or cropping. It can move food onto different plates or backgrounds, and even modify the food itself - making portions look bigger or digitally filling in gaps for a more polished look.
This approach goes further than traditional retouching or generic stock photos. The AI is capable of generating convincing images of dishes that, in some cases, never actually existed in this form.
Cohere's new Command A Vision model is designed to handle images, diagrams, PDFs, and other types of visual data. Cohere says the model outperforms GPT-4.1, Llama 4 Maverick, Pixtral Large, and Mistral Medium 3 on standard vision benchmarks.
The model's OCR can recognize both the text and the structure of documents such as invoices and forms, outputting the extracted data in structured JSON. Command A Vision can also process real-world images, like identifying potential risks in industrial environments, the company says.
Command A Vision is available through the Cohere platform and for research on Hugging Face. The model can run locally with either two A100 GPUs or a single H100 using 4-bit quantization.
