#12 The first LLM that has a bigger IQ score than the average human
The singularity is coming closer by the day. Anthropic launched its new LLM Claude 3 which has a IQ of 101, more than the average human. In this Substack we’ll take a peak at their benchmark scores. Also we’ll check in what investments are being made in the world of AI robotics, learn about audio control by Adobe and announce a new AI feature in our PresentersWall tool. Let’s go!
Anthropic launches Claude 3, a better LLM-model than GPT-4 (well, sort of)
Anthropic has introduced the Claude 3 model family, which consists of three highly capable models: Claude 3 Haiku, Claude 3 Sonnet, and Claude 3 Opus. Each successive model offers increasing levels of intelligence and capability, with the main focus being on the Opus model. According to the benchmarks released by Anthropic, Claude 3 appears to outperform GPT-4 at launch, suggesting that it could be a real competitor to GPT-4 and Google Gemini 1.5 Pro.
The benchmark scores released by Anthropic, the founder of Claude 3, are displayed in the image above. A higher number indicates a better score.
Opus and Sonnet are now available in the Claude API, while Haiku will be available soon. This means that users can try it out for themselves by registering an account on https://console.anthropic.com/ and accessing the console tab. At least this is the way to test it when you are located in Europe. Sadly it's not yet available in the EU regularly yet.
The models are designed to make it easier for users to adhere to brand guidelines and output structured formats. Anthropic has stated that they will continue to enhance the models' intelligence and enterprise features in the coming months. Although it’s already proved to be pretty capable and have an IQ of 101 (which is the first time it exceeds the average human in intelligence. ChatGPT-4 has an IQ of 85).
Our own initial experiences with the API have been very positive. As Anthropic themselves state, the API appears to be more consistent with your preferred tone of voice, and it provides responses that are less reminiscent of the typical "ChatGPT" style. Instead, it generates copy that is more practical, realistic, and ultimately more usable.
Figure.ai found investments for it’s AI robots
In an exciting development for the field of robotics, humanoid robot startup Figure AI has raised $675 million at a $2.6 billion valuation from a star-studded group of investors, including Jeff Bezos, Nvidia, Microsoft, and OpenAI. The funding will be used to accelerate the development and commercial deployment of Figure's general-purpose humanoid robot, Figure 01, which is designed to perform everyday tasks autonomously in industries facing labor shortages, such as manufacturing, shipping, logistics, warehousing, and retail.
As part of the agreement, Figure will collaborate with OpenAI to incorporate the company's latest AI models into their robots, significantly improving their natural language processing abilities. This partnership represents an exciting development in the field of humanoid robotics, where Figure faces strong competition from industry leaders such as Boston Dynamics and Tesla.
Adobe previews new cutting-edge generative AI tools for crafting and editing custom audio
New experimental work from Adobe Research is set to change how people create and edit custom audio and music. An early-stage generative AI music generation and editing tool, Project Music GenAI Control allows creators to generate music from text prompts, and then have fine-grained control to edit that audio for their precise needs.
“With Project Music GenAI Control, generative AI becomes your co-creator. It helps people craft music for their projects, whether they’re broadcasters, or podcasters, or anyone else who needs audio that’s just the right mood, tone, and length,” says Nicholas Bryan, Senior Research Scientist at Adobe Research and one of the creators of the technologies.
A Touch of Magic in the PresentersWall Tool with Generative AI: From Wordcloud to Image
The AI innovations are flying around you in 2024, with one tool rapidly succeeding another. To transform the world of event interaction, the PresentersWall team proudly introduces its latest generative AI application: Wordcloud-to-Image. This feature not only adds an innovative layer to audience interaction during events but also increases the fun factor.
Why? You'll read everything about it in this article: https://www.livewall.co/blogs/presenterswall-generative-ai-wordcloud-to-image