pragma.vision Your technology observatory

Verification register Energy, Climate, Space & Materials

Readiness verdict

Jua EPT-2

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
85

Public ambition and stated capability

Reported
73

Observed practitioner reporting

Verified
59

Independently supported evidence

Gap
+26

Claimed minus verified

Evidence strength Strong

Decision

What the current evidence supports

Human editorial judgment · 2026-06-28

Proceed with caution

Why
The cost/cadence story (up to 24 updates/day, ~0.25 kWh/inference, $0.20-$15/run) is genuinely attractive and the API is live, but the accuracy claims are entirely self-reported and independent literature flags exactly the extreme-event weakness that would hurt energy use cases. A credible operational open baseline (AIFS-ENS) now exists to benchmark against for free.
Next
Run a head-to-head shadow trial of EPT-2 vs an open baseline (ECMWF AIFS-ENS, now operational since 1 Jul 2025, and/or GraphCast) on the exact variables and lead times we need, scoring extreme-event days separately; treat all vendor RMSE/CRPS numbers as a hypothesis to verify.

Constraints

Blockers

No named blocker is present in the current public projection.

Evidence summary

Derived counts

AI-assisted assembly

Total
7
Tier 1
5
Tier 2
1
Tier 3
1
Supports
4
Contradicts
2
Context
1
Latest observed
2026-01-01

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.