Data lakehouse provider Databricks has unveiled a new large language model (LLM) training method, TAO that will allow enterprises to train models without labeling data.

Databricks has achieved performance gains with AMD GPUs, recording a 1.13x improvement in training performance when using ROCm 5.7 and FlashAttention-2 compared to previous results with ROCm 5.4 ...

Databricks acquired MosaicML for $1.3 billion and is steadily rolling out tools that help developers create A I apps rapidly. The Mosaic research team at Databricks developed the new TAO method.