pragma.vision Your technology observatory

Verification register Compute & Web Infra

Readiness verdict

NVIDIA GB300 NVL72

A dated reading of what is claimed, reported, and independently verified in the current evidence.

As of
2026-06-28
Revision
1
Method
v1.0.0

Current reading

The readiness gap, in one scan

AI-assisted assembly · derived results

Claimed
90

Public ambition and stated capability

Reported
86

Observed practitioner reporting

Verified
79

Independently supported evidence

Gap
+11

Claimed minus verified

Evidence strength Strong

Decision

What the current evidence supports

Human editorial judgment · 2026-06-28

Adopt with guardrails

Why
Officially shipping with first real hyperscaler deployments and the strongest FP4 reasoning-inference profile available, but power/cooling intensity and early-ramp supply mean adoption is rent-via-cloud with capacity and facility guardrails, not bare-metal for most.
Next
Consume via CoreWeave/Dell-deployed instances rather than self-hosting; size around the ~120 kW/rack liquid-cooling envelope and the 2H-2025 partner ramp; benchmark real FP4 inference throughput before committing reserved capacity.

Constraints

Blockers

No named blocker is present in the current public projection.

Evidence summary

Derived counts

AI-assisted assembly

Total
6
Tier 1
0
Tier 2
4
Tier 3
2
Supports
3
Contradicts
1
Context
2
Latest observed
2025-09-01

Counts and dates only. Raw signals, private excerpts, trust records, and internal corpus material are not published here.

Publication record

Revisions

Initial public reading

This is the initial public reading. No earlier readiness change is recorded.

Your opinion

Tell us anything.

What works, what doesn't, what's missing — especially about our watches, lenses, and the register itself. Anonymous is fine; leave an email if you'd like a reply.