That’s why, if ever in doubt, it’s better to ask somebody about a meme’s true meaning. Did I say “somebody”? Oops, I mean something.
Something like Grok, Elon Musk’s xAI creation. Now, Grok has a new version: Grok-1.5 Vision.
Mr. SpaceX/Neuralink/Tesla recently launched Grok-1.5 – that was shortly after Google unveiled Gemini 1.5 last month. Although Musk’s xAI claimed that Grok-1.5 LLM approaches the performance of GPT-4, it lacks multimodal capability. However, xAI’s newly announced Grok-1.5 Vision overcomes this limitation by seamlessly processing both textual and visual information (via MSPowerUser).
Connecting the digital and physical worlds
Grok-1.5 Vision (a.k.a. Grok-1.5V) represents xAI’s inaugural multimodal model designed to bridge the gap between digital and physical realms. According to a recent blog post by the company, Grok excels in benchmark testings, demonstrating superior real-world spatial understanding. Additionally, Grok-1.5V has the capability to interpret a diverse range of visual data, encompassing documents, diagrams, charts, screenshots, and photographs.
Some notable applications of Grok-1.5V include generating code from diagrams, calculating calorie counts, crafting bedtime stories based on drawings, or helping meme aficionados get a better understanding of memes.
xAI asserts that Grok-1.5V surpasses the competition’s latest AI creations, like Google’s Gemini Pro, in benchmark testings.
While Grok-1.5V is not currently available, xAI plans to release it soon for early testers and existing Grok users as a preview.