How good is Runway Gen-3?
How good is the new Runway Gen-3 video AI really? One Reddit user treated himself to a year of access and let others decide which prompts to try. See the results here.
SCR
Anthropic helps to craft better prompts
Anthropic has released new features for its Claude language model to help developers create better prompts to optimize the performance of their AI applications.
Microsoft MInference increases the speed of LLMs
Microsoft’s new “MInference” technology promises to significantly increase the processing speed of large language models by reducing the preprocessing time of long texts by up to 90%. An interactive demo on Hugging Face allows developers to test the technology and explore its capabilities.
Poe adds interactive applications
Quora’s AI chatbot platform, Poe, introduced Previews, a new feature that allows users to create interactive applications in chats. These include data visualizations, games, or even drum machines that access information from uploaded files. The created applications can be shared via a link.
Salesforce xLAM-1B announced
Salesforce has developed xLAM-1B, a small but powerful AI model that supposedly outperforms larger models from OpenAI and Anthropic in function calls.
ElevenLabs AI Voice Isolator introduced
ElevenLabs has introduced a new free service called AI Voice Isolator, which removes unwanted background noise from movies, podcasts or YouTube videos. Unlike other programs that can only remove constant noise, the Voice Isolator also handles irregular noises such as a door opening or someone clapping.
ElevenLabs Reader reads any text aloud for you
ElevenLabs has released a new app called Reader, which allows users to have any text read aloud in AI voices. New are “Iconic Voices”, which recreate the voices of deceased stars such as Judy Garland, James Dean and Laurence Olivier. The company acquired the rights to the voices from CMG Worldwide and stresses that the …
Apple 4M is a multimodal powerhouse
The “4M” AI model provides a glimpse into Apple’s progress in artificial intelligence. Developed in collaboration with EPF Lausanne, the model can convert text to images, recognize objects, and manipulate 3D scenes based on speech input.
Resemble Detect-2B helps to recognize audio deepfakes
Resemble AI has introduced Detect-2B, a new audio deepfake detection model that claims to have 94% accuracy. The model looks for subtle artifacts to determine whether speech is real or artificially generated.
AWS introduces App Studio
With the new AWS App Studio, organizations are supposed to be able to build scalable internal applications in minutes-without any programming skills. Using generative AI, App Studio uses natural language descriptions to create applications that can be integrated with internal systems. According to Amazon, the tool handles all aspects of application development, from deployment to …