NSF's America's Seed Fund is tech-agnostic, takes zero equity, and lets you propose your own idea. The gate is the Project Pitch: a short online pre-screen (no registrations required — you can send it before you've incorporated), capped at two pitches per company per year. Roughly three in four pitches are declined, but an invite earns you the right to submit a full proposal for a Phase I up to ~$305,000 over 6–18 months; Phase II runs to $1.25M plus supplements. After the 2026 restart, full-proposal deadlines landed July 27 and November 4, 2026, settling into a twice-yearly rhythm. The bar to clear: genuine technical risk. "We integrated existing parts" doesn't pitch well; "nobody knows if this physics closes" does.
NASA runs contracts against topics tied to its mission directorates and centers. The big 2026 change: the single annual solicitation is gone, replaced by a rolling omnibus BAA (April 17, 2026 through September 30, 2027) with appendices that open and close in roughly four-to-five-week windows through the year. Phase I runs about $150–225K over six months (the separate, commercially-focused SBIR Ignite track sits near $150K); Phase II has recently been in the $850K range — verify the live appendix, as published caps have moved. NASA also has the richest post-Phase-II ladder (Phase II-E and II-X matching extensions). If your tech maps to a NASA center's roadmap — propulsion, entry heating, cryogenics, in-space assembly — this is your lane, but you have to watch the appendix calendar like a hawk.
DOE awards grants (Phase I $200–250K, Phase II commonly quoted up to ~$1.5M+) against topics written by program offices and national-lab scientists — off-topic proposals are administratively rejected, so fit is everything. Two releases a year in normal times; the compressed 2026 cycle had Release 2 applications due July 7, 2026. One famous quirk: DOE has required a short letter of intent weeks before the application — miss the LOI, and you cannot apply (reports suggest DOE may drop this requirement; check the live funding opportunity). Fusion, grid hardware, power electronics, carbon capture, advanced manufacturing: if it plugs into the energy system, DOE probably has a topic for it.
All three agencies fund physics-heavy work — propulsion, thermal, materials, grid hardware — where the Phase I deliverable is largely modeling and analysis. If your concept needs CFD, FEA or electromagnetics evidence before reviewers will believe it, ask us what an Ansys evaluation could cover while you're still pre-award.
See if you qualify for an Ansys eval + engineering support at no cost How access works →| NSF | NASA | DOE | |
|---|---|---|---|
| Type | Grant | Contract | Grant |
| Phase I | ~$305K / 6–18 mo | ~$150–225K / 6 mo | $200–250K / 6–12 mo |
| Gate | Project Pitch (mandatory) | Appendix windows (~4–5 wks) | Letter of intent (historically) |
| Best for | Broad deep-tech, first-timers | Space/aero tech on NASA roadmaps | Energy, grid, materials |
All three can run in parallel with a DoD application, as long as you disclose overlaps — see grants vs. contracts for how the cultures differ day-to-day.
Official sources for this guide: NSF 26-510 solicitation · NASA SBIR/STTR PY2026 hub · DOE SBIR (Office of Science). Figures change; always confirm on the official page before relying on them.
DSIP, monthly releases, AFWERX open topics, Navy base+option, Army xTech — how the $2B+ program actually works.
Read →Grant culture, waiver-boosted budgets, three annual deadlines, and why FDA's simulation guidance changes the game.
Read →