1,000+ Opportunities
Find the right grant
Search federal, foundation, and corporate grants with AI — or browse by agency, topic, and state.
The NIH Common Fund launched PRIMED-AI, Precision Medicine with AI: Integrating Imaging with Multimodal Data, to combine medical imaging with other health data in order to develop AI-powered clinical decision support tools that advance personalised medicine, and announced five funding opportunities with staggered deadlines.
RFA-RM-27-014 funds a Validation Center to independently verify and assess AI tools developed through the programme, due 2 October. RFA-RM-27-015 funds a Logistics Center to coordinate activities and provide administrative and outreach support, due 2 October. RFA-RM-27-011 funds a Multi-use Frameworks Playbook to develop standard processes for responsible AI use and data management, due 9 October.
RFA-RM-27-012, Data-to-Model Academic-Industrial Partnerships, supports teams integrating multimodal data and developing clinical decision support tools, due 19 October 2026. RFA-RM-27-013, Model-to-Clinic, moves validated tools into real-world clinical settings with eight awards expected, due 19 October 2026.
The architecture is the notable element and distinguishes PRIMED-AI from conventional AI-in-imaging funding: NIH is funding an independent Validation Center alongside the model-development awards, meaning tools built under the programme will be assessed by a separate team rather than self-reported, and a responsible-AI and data-management playbook is funded as a first-class deliverable rather than as a compliance appendix.
Applicants should expect validation standards and data governance to be externally imposed rather than self-defined. All five use UG3/UH3 or similar cooperative agreement mechanisms with substantive NIH staff involvement. Eligibility is broad, including foreign organisations and international collaborations.
The recorded deadline of 19 October 2026 is the latest of the five.
Get a weekly digest of new grants like this
A free weekly digest of new foundation and federal funding opportunities as they're added to Granted. Unsubscribe anytime.
Or search similar grants →According to the current listing, eligibility includes: Eligibility across the PRIMED-AI opportunities is broad by NIH standards and includes nonprofits with and without 501(c)(3) status, higher education institutions both public and private as well as independent school districts, government entities at state, county, city, tribal and special district level, for-profit organisations including small businesses, and other organisations; foreign organisations and international collaborations are eligible, which is not true of every Common Fund initiative and matters for consortia assembling imaging cohorts across countries. The real filter is fit to a specific opportunity, and the five are quite different in what they require. The Validation Center (RFA-RM-27-014) requires demonstrable independence from the tool developers and expertise in rigorous model evaluation, so a team that also intends to build models under the programme has a conflict to resolve. The Logistics Center (RFA-RM-27-015) is a coordination and administrative capability award, not a research award. The Multi-use Frameworks Playbook (RFA-RM-27-011) requires expertise in responsible AI governance and health data management rather than in model architecture. Data-to-Model Academic-Industrial Partnerships (RFA-RM-27-012) requires a genuine academic-industrial partnership, which means intellectual property, data rights and publication terms must be settled with an industry partner before submission. Model-to-Clinic (RFA-RM-27-013) aims to catalyse translation of AI-enabled, image-centered, multimodal clinical decision support tools from prototype toward clinical implementation, requires that projects address unmet health challenges in precision medicine with demonstrated potential for improved patient outcomes and healthcare processes, and anticipates eight awards; it therefore requires clinical site access and an implementation plan, not merely a validated model. All use cooperative agreement mechanisms with substantive NIH programme involvement and milestone-gated phase transitions. Applicants should read the individual notices on grants.nih.gov for budget guidance, due dates and specific requirements, and may contact ODPRIMED-AI@od.nih.gov. Confirm the full requirements in the official notice before applying.
The current listing shows NIH has not published award ceilings or floors in the summary materials for the five PRIMED-AI funding opportunities, which is the most consequential planning gap and should be treated as such: the figures recorded here, amount_min 1,500,000 and amount_max 6,000,000 US dollars, are not published programme amounts but an estimate based on typical NIH Common Fund UG3/UH3 cooperative agreement scales, and applicants must obtain actual budget guidance from the individual notices of funding opportunity on grants.nih.gov or from ODPRIMED-AI@od.nih.gov before building a budget. Several structural features bear directly on cost. The UG3/UH3 mechanism is a two-phase cooperative agreement: the UG3 phase funds a shorter milestone-driven development period and transition to the larger UH3 phase is contingent on meeting those milestones, so the budget must be constructed as two defensible phases rather than one continuous project, and the UG3 milestones effectively become the contractual basis for continued funding. Cooperative agreements also entail substantive NIH programme staff involvement in project direction, which has real administrative cost and reduces autonomy relative to an R01. The specific cost drivers in this programme are imaging-centric: assembling and harmonising multimodal datasets that pair medical imaging with clinical, genomic and other health data; de-identification and data-use agreements across institutions; and for the Model-to-Clinic opportunity, the cost of clinical implementation including site onboarding, clinician workflow integration, prospective evaluation and regulatory navigation, which routinely exceeds model development cost and is the line item most often underestimated. The Data-to-Model opportunity requires an academic-industrial partnership, so intellectual property, data rights and publication terms must be negotiated with the industry partner before submission, a process that takes months. Eight awards are expected under Model-to-Clinic (RFA-RM-27-013). Verify award ceilings, matching requirements, and allowable costs in the official notice.
The published deadline was October 19, 2026, which has passed. Check the official notice for any future application windows before investing time in a proposal.
NIH Common Fund PRIMED-AI Precision Medicine with AI Integrating Imaging with Multimodal Data Program Suite Including Data-to-Model Model-to-Clinic Validation Logistics and Frameworks Playbook is funded by National Institutes of Health (NIH) Common Fund. Verify program details on the funder's official page before applying.
Start from the official opportunity page linked in this listing — it carries the sponsor's submission instructions.
RFA-RM-27-011 funds the development and testing of a multi-use frameworks playbook for precision medicine with AI, within the NIH Common Fund's PRIMED-AI programme. Where other PRIMED-AI awards build tools, validate them or move them toward the clinic, this one produces the shared methodology: a reusable framework that other teams can apply when integrating medical imaging with multimodal health data to build AI-based clinical decision support. The premise is that the field repeatedly re-solves the same design problems, and that codifying frameworks for data harmonisation, model development, evaluation and clinical integration is a public good worth funding separately. Funding is up to 300,000 US dollars per year for two years, which makes this the most accessible entry point into a major NIH AI health programme for groups with strong methodological credentials but without the institutional scale to host a centre. Crucially, the deliverable is to be developed and tested, not merely written, so proposals need a concrete validation plan showing the playbook works when applied by teams other than its authors. The RFA was released on 30 June 2026 with applications due 9 October 2026, the second of the five PRIMED-AI deadlines. For informatics methodologists, implementation scientists and groups working on AI reporting and reproducibility standards, this is a rare opportunity in which methodological synthesis is the funded product rather than a byproduct.
Model-to-Clinic, RFA-RM-27-013, is the translational arm of the NIH Common Fund's PRIMED-AI programme. It funds teams to take AI-enabled, image-based multimodal clinical decision support models and move them toward actual use in clinical settings, addressing the gap that has defined clinical AI for a decade: models that perform well retrospectively and then fail to change practice. The award uses a two-phase UG3/UH3 cooperative agreement, with a milestone-gated preparatory phase capped at 450,000 US dollars in direct costs per year followed, on successful transition, by a phase capped at 1,000,000 US dollars in direct costs. Clinical trials are optional, so teams can pursue prospective evaluation or focus on workflow integration, interoperability and clinician-facing deployment without being forced into a trial design. PRIMED-AI deliberately splits the pipeline across five opportunities, and M2C sits downstream of the Data-to-Model academic-industrial partnerships and alongside an independent Validation Center that will evaluate tools the programme produces, so successful applicants should expect their models to be externally scrutinised rather than self-assessed. The RFA was released on 30 June 2026 with applications due 19 October 2026. For academic medical centres with both AI development capability and real clinical implementation reach, this is one of the few federal mechanisms that pays specifically for the translation step rather than treating it as dissemination.
The FY2026 Emerging Health Innovators Initiative, Notice ID ARPA-H-SOL-26-150, is ARPA-H's dedicated funding route for early-career principal investigators and carries two explicitly AI-focused topics among its 16 anticipated topic areas: Causally-Grounded Alignment for Clinical Generative Artificial Intelligence Tools, and Protecting Patient Data in Life Science AI. The first targets the core failure mode of clinical generative AI, which is that models aligned on correlational patterns in clinical text produce fluent recommendations that are not causally grounded in the underlying disease mechanism; the topic asks for alignment approaches anchored in causal structure rather than human preference ratings. The second addresses privacy-preserving computation over life-science datasets, where the tension between model utility and patient re-identification risk currently blocks multi-institution training. The structure is two-stage and the first gate is the binding one: a required Solution Summary is due 30 October 2026 at 2:00 PM ET, and only proposers who clear that stage are positioned for the full proposal deadline of 10 February 2027 at 2:00 PM ET. An optional virtual Proposers' Day was held on 13 October 2026. ARPA-H anticipates 20 or more awards per year at up to 350,000 dollars per year for two years with an optional third year. Submission is through the ARPA-H Solutions portal at solutions.arpa-h.gov after retrieving the solicitation and its attachments from SAM.gov under the notice ID.
NOT-RM-26-004 asks the public for NIH-wide research challenges worth a 10-year, cross-institute program. There is no budget, no biosketch, no page limit, and no award. What there is: the five criteria every Common Fund investment has met since 2006, a new bullet about small-lab burden, and a fund whose FY 2026 request came in at $347.4 million — down 49.3 percent.
Read articleThe NIH Common Fund launched PRIMED-AI — Precision Medicine with AI: Integrating Imaging with Multimodal Data — as five coordinated funding opportunities (RFA-RM-27-011 through -015) that build a full pipeline from standards to clinic. Here is how the Playbook, Data-to-Model partnerships, Model-to-Clinic translation, Validation Center, and Logistics Center fit together, what each pays, who is eligible, and how to position before the October 2026 deadlines.
Read articleAvoid common NIH grant proposal mistakes including vague specific aims, weak methodology, and poor budget justification that lead to rejection.
Read article