1,000+ Opportunities
Find the right grant
Search federal, foundation, and corporate grants with AI — or browse by agency, topic, and state.
DARPA's CLARA (Compositional Learning-And-Reasoning for AI Complex Systems Engineering) program is a fundamental research initiative seeking to tightly integrate automated reasoning (AR) with machine learning (ML) to create high-assurance AI systems that are demonstrably trustworthy, not just empirically good.
Published under solicitation DARPA-PA-25-07-02 on February 10, 2026, the program explicitly criticizes the current industry approach of tacking automated reasoning onto large language models and instead demands tight compositional integration of ML and AR components with hierarchical structure and transparent operation.
CLARA is structured in two technical areas: TA1 funds development of new high-assurance ML/AR composition approaches including theory, algorithms, and open-source code, while TA2 builds a software composition library integrating validated TA1 tools into a common framework. All software must be released as open source under Apache 2. 0 license.
Application domains include autonomous systems, command and control, kill web operations, supply chain logistics, wargaming, and medical, financial, and legal domains. The program targets composition of Bayesian methods, neural networks, and logic programs. DARPA aims to execute awards by June 9, 2026, within 120 days of posting.
Get alerted about grants like this
Save a search for “DARPA Defense Sciences Office” or related topics and get emailed when new opportunities appear.
Search similar grants →Based on current listing details, eligibility includes: Open to universities, research organizations, small businesses, and defense contractors with expertise in machine learning combined with formal methods, including probabilistic programming, automated theorem proving, satisfiability solvers, type-theoretic verification approaches, or neurosymbolic AI architectures. U.S. and allied nation organizations eligible per DARPA standard terms. Applicants should confirm final requirements in the official notice before submission.
Current published award information indicates Up to $2 million per award over a 24-month period of performance. The program funds two technical areas: TA1 for development of high-assurance ML/automated reasoning composition approaches, and TA2 for building a software composition library integrating TA1 tools. Always verify allowable costs, matching requirements, and funding caps directly in the sponsor documentation.
The current target date is April 17, 2026. Build your timeline backwards from this date to cover registrations, approvals, attachments, and final submission checks.
Federal grant success rates typically range from 10-30%, varying by agency and program. Build a strong proposal with clear objectives, measurable outcomes, and a well-justified budget to improve your chances.
Requirements vary by sponsor, but typically include a project narrative, budget justification, organizational capability statement, and key personnel CVs. Check the official notice for the complete list of required attachments.
Yes — AI tools like Granted can help research funders, draft proposal sections, and check compliance. However, always review and customize AI-generated content to reflect your organization's unique strengths and the specific requirements of the solicitation.
Review timelines vary by funder. Federal agencies typically take 3-6 months from submission to award notification. Foundation grants may be faster, often 1-3 months. Check the program's timeline in the official solicitation for specific dates.
Many federal programs offer multi-year funding or allow competitive renewals. Check the official solicitation for continuation and renewal policies. Non-competing continuation applications are common for multi-year awards.