Meta

Meta's latest open-source model offering performance comparable to Llama 3.1 405B at a fraction of the cost. Excellent for self-hosting.

Multimodal model with vision capabilities available in 11B and 90B parameter sizes. Supports image understanding and reasoning.

Meta's largest open-source model with 405 billion parameters. Competitive with leading closed models on benchmarks.

MetaSeq is Meta's internal library for training large-scale sequence models, used to train models like OPT (Open Pre-trained Transformer). It provides the infrastructure for efficient distributed training of language models with billions of parameters.