Musk's artificial intelligence company xA Launched in late March Grok-1.5 After the large language model,Recently launched the firstMultimodal Model Grok-1.5 Vision.
xAI said it will soon invite early testers and existing Grok users to test Grok-1.5 Vision (Grok-1.5V), which can not only understand text, but also process the content in documents, charts, screenshots and photos.
xAI said: "Grok-1.5V is comparable to existing cutting-edge multimodal models in many areas such as multidisciplinary reasoning, document understanding, scientific graphs, table processing, screenshots and photos."
In its official press release, xAI demonstrated seven Grok-1.5V cases, including converting flowchart sketches on a whiteboard into Python code, generating bedtime stories based on children's drawings, interpreting buzzwords, converting tables into CSV file format, and more.
xAI also shared the running results of Grok-1.5V, which outperformed mainstream competitors such as GPT-4V, Claude 3Sonnet, Claude 3 Opus and Gemini Pro 1.5 in the RealWorldQA benchmark.