MachineLearningLM/MachineLearningLM-7B-v1

MachineLearningLM/MachineLearningLM-7B-v1 is a 7.6 billion parameter language model continuously pretrained on millions of synthetic tabular machine learning tasks. Developed by MachineLearningLM, it is designed for robust many-shot in-context learning, scaling effectively from 8 to 1,024 examples. This model demonstrates significant improvement on unseen tabular tasks and achieves Random-Forest-level numerical modeling robustness, making it suitable for complex data analysis and prediction challenges.