IBM announces availability of open-source Mistral AI model on watsonx platform

IBM announced the availability of the open-source Mixtral-8x7B large language model – LLM -, developed by Mistral AI, on its watsonx AI and data platform. IBM offers an optimized version of Mixtral-8x7B that, in internal testing, was able to increase throughput – or the amount of data that can be processed in a given time period – by 50% when compared to the regular model. This could potentially cut latency by 35-75%, depending on batch size – speeding time to insights. Mixtral-8x7B was built using a combination of Sparse modeling – a technique that finds and uses only the most essential parts of data to create more efficient models – and the Mixture-of-Experts technique, which combines different models that specialize in and solve different parts of a problem.