A truly open large language model released, developed with LUMI

The Allen Institute for AI (AI2) has released OLMo 7B, a truly open, state-of-the-art large language model alongside the pre-training data and training code. This empowers researchers and developers to use the best and open models to collectively advance the science of language models.

– We are happy that we can contribute to this important initiative by providing the computing capacity from the LUMI supercomputer along with our expertise. Public supercomputers like LUMI play a vital role in open and transparent AI infrastructure, said Dr. Pekka Manninen, Director of Science and Technology, CSC, in Allen Institute for AI’s blog.

The LUMI supercomputer in Finland is hosted by CSC and owned by EuroHPC Joint Undertaking and 10 European countries. It is the fastest supercomputer in Europe and is known for its entirely carbon-free operations. LUMI was critical in supporting the pre-training work necessary to develop OLMo.

Read more about OLMo 7B in AI2’s blog.

Read also: LUMI part of AI2 OLMo, an open language model made by scientists, for scientists