https://unitedstatesofamericasuperintelligence.com/companies/stanford-university/ Dated, sourced news items in the USASI catalog that mention Stanford University. Independent project. Not a United States government website. en-us Tue, 13 Oct 2026 12:00:00 GMT https://unitedstatesofamericasuperintelligence.com/news/dspy-3-4-1-makes-instructionproposer-gepa-default/ https://unitedstatesofamericasuperintelligence.com/news/dspy-3-4-1-makes-instructionproposer-gepa-default/ Tue, 13 Oct 2026 12:00:00 GMT release On October 12, 2026 DSPy 3.4.1 was published on the project's GitHub repository. The release notes say it makes GEPA's structured InstructionProposer the default instruction proposer, lets dspy.RLM build DSPy sub-agents from inside its REPL, adds Python 3.15 support, and fixes a 3.4.0 import crash. The notes warn that GEPA's default reflection request changes shape, so optimization results can differ from 3.4.0 even with the same seed, and that the reflection LM must now be a dspy.LM that can return JSON. USASI has not tested this release. https://unitedstatesofamericasuperintelligence.com/news/marin-begins-pretraining-535b-a23b-model/ https://unitedstatesofamericasuperintelligence.com/news/marin-begins-pretraining-535b-a23b-model/ Tue, 13 Oct 2026 12:00:00 GMT research In a launch note dated September 3, 2026, first published on the Open Athena blog and mirrored by Marin, Marin said it had started pretraining Marin 535B-A23B a couple of weeks earlier. The note describes a mixture-of-experts model with 535 billion total parameters and 23 billion active, planned to pretrain for about three months on 18 trillion tokens, and says the team ported its infrastructure and training stack from Google TPUs on GCP to NVIDIA GPUs on CoreWeave before launch, with post-training to follow. It says the entire run, including architecture, data, and predicted loss, is documented on GitHub; it does not give a date or license for releasing the model's weights. https://unitedstatesofamericasuperintelligence.com/news/terminal-bench-4-0-released/ https://unitedstatesofamericasuperintelligence.com/news/terminal-bench-4-0-released/ Tue, 29 Sep 2026 12:00:00 GMT release The Terminal-Bench team published version 4.0 of its command-line agent benchmark on August 28, 2026, a month after announcing that it would be maintained as a continuous benchmark with semantic versioning. The release removed eight tasks, fixed nineteen, and set a flat eight-hour agent timeout for every task; it is a major version because these changes require re-running trials. It runs through the Harbor framework.