Macaw (Multi-angle c(q)uestion answering) is a ready-to-use model capable of general question answering, showing robustness outside the domains it was trained on. It has been trained in "multi-angle" fashion, which means it can handle a flexible set of input and output "slots" (like question, answer, explanation) .

Macaw was built on top of T5 and comes in different sizes: macaw-11b, macaw-3b, and macaw-large, as well as an answer-focused version featured on various leaderboards: macaw-answer-11b (see below).

Even though Macaw is an order of magnitude smaller than GPT-3 (using just 11 billion parameters to GPT-3’s 175 billion parameters), it outperforms the popular model, making it a very promising evolution in the commoditization of powerful language models.

Macaw can successfully tackle a broad range of question types including general knowledge, meta reasoning, hypothetical, and story understanding. AI2’s Aristo team developed Macaw as part of their efforts toward building AI systems that can read, reason, and explain their answers.