Spanish startup Multiverse Computing has successfully raised €189 million in a Series B funding round, driven by its innovative technology, CompactifAI. This quantum-computing inspired compression technique can reduce the size of large language models (LLMs) by up to 95% without sacrificing performance. Their compressed models, available for AWS and on-premises use, promise significant speed improvements and cost reductions for businesses. Notably, Multiverse's models can even run on low-power devices like PCs and Raspberry PIs, marking a significant advancement in accessibility for AI technologies. The company is co-founded by Román Orús, noted for his work in computational tools mimicking quantum systems.
CompactifAI is a quantum-computing inspired compression technology that reduces the size of LLMs by up to 95% without impacting model performance, potentially revolutionizing AI.
With Multiverse's slim models available on AWS, companies can experience a reduction of 50%-80% in inference costs while also improving processing speed by 4-12 times.
Collection
[
|
...
]