TriviaQA: Results as Published

TriviaQA tests reading comprehension and recall with about 95,000 trivia questions written by enthusiasts, each paired with evidence documents. Model cards usually print exact-match accuracy on the answers, often without the evidence documents, which turns it into a knowledge test.

Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.