Hardware giant Nvidia has launched a new open source AI model, NVLM 1.0. The LLM is based on a different version with 72 billion parameters.

Nvidia’s new multimodal LLM (Large Language Model) delivers strong performance on complex vision language tasks. In terms of text, NVLM 1.0 does not take first place in AI benchmarks. The model processes images very well. The researchers used examples to show that the model can analyze images and solve complex mathematical problems step by step.

After multimodal training, performance improves on text prompts while deteriorating on many other models.

New chapter in AI development with an open model

According to Venturebeat, they want to compete with market leaders with closed models such as GPT-4o from OpenAI or Claude 3.5 from Anthropic. By making Nvidia public its parameters (model weights) and soon the training codes, researchers and developers will have full access to advanced technology. Smaller organizations or companies can conduct research in the field of AI more efficiently. It may not be used commercially.

Nvidia’s decision to open everything up could significantly accelerate AI research. It is also hoped that competition will be forced to be completely transparent. That would only benefit innovation and collaboration.

On the other hand, with so much openness you can also ask questions. Such powerful and, above all, accessible AI can quickly fall into the wrong hands, and that raises ethical questions. The impact of Nvidia’s decision will become clear in the coming weeks and months.