The document puts $1.8 billion into the Virtual Biology Initiative. That number does not prove a working virtual cell, a named model architecture, a benchmark score, or a ship date. It is a funding and data-generation commitment, not a model release.

The split shows where the constraint sits. DOE brings more than $500 million over five years, NIH brings prior federal investment above $500 million, and Google DeepMind, Isomorphic Labs, and Meta add $300 million. Biohub's $500 million is split into $400 million for measurement tools and $100 million for outside research. The buyer or fab should read that as a bottleneck in instruments, standards, and data throughput, not in announced compute.

What remains unmeasured is the model itself. The release names no architecture, benchmark, parameter count, context length, dataset size, or working virtual cell. It also leaves the timing open: the initiative was announced in April 2026, but no model release date is stated.

The money is weighted toward measurement and existing federal data, not toward a released model. For the buyer or fab, the binding constraint is instrument throughput and data standards, not announced compute.

The document leaves out model architecture, benchmark, parameter count, context length, dataset size, and a working virtual cell.

The funding split points to instrument capacity, dataset standardization, and data throughput as the limiting factors, not to a shipped model or benchmark.