Databricks reported that Stanford's team won its June 17 Grounded Reasoning Cup with 63.3% accuracy on OfficeQA Pro V2, a held-out benchmark built from about 120,000 pages of U.S. Treasury documents, …
Mixedbread shipped Toast 1, a specialized search agent that matches Claude Opus 5 and GPT-5.6 Sol on deep-search benchmarks while running up to 10× cheaper and 12× faster, according to vendor-run test…
Mixedbread AI released Toast 1, a specialized search agent that matches or outperforms Claude Opus 5 and GPT-5.6 Sol while being up to 10× cheaper and 12× faster. In Databricks' OfficeQA Pro V2 benchm…
Databricks released OfficeQA Pro V2, a benchmark with 90 questions grounded in about 120,000 pages of U.S. Treasury data, finding that out-of-the-box frontier agents averaged 37.5% accuracy, while age…