‹ All rumors

Did Xplore release RWE-bench showing frontier models pass only about 37% of real engineering tasks?

ConfirmedLast checked
XploreFirst seen

Verdict

Confirmed. Xplore’s own RWE-bench page describes twenty real engineering tasks with physics verifiers and early ~37% top pass rate. DailyCited lists this as Official on the Oct 28 brief.

Evidence timeline · 1 source · newest first

  1. Unconfirmed → Confirmed

  2. Confirmed

    Xplore released RWE-bench, twenty mechanical, electrical and aerospace tasks written by practicing engineers and graded by deterministic physics verifiers rather than LLM judges. Early leaderboard figures put the best model at about 37% pass rate (GPT-6 Astra cited in secondary coverage), with crankshaft CAD, DDR3 routing and turbine-vane CFD among the tasks. DailyCited lists this as Official on the Oct 28 brief.

    Source: Xplore · Official story →
Quick answers
Is Xplore RWE-bench real-world engineering AI benchmark out yet?
Yes. Xplore confirmed it on Oct 27.
On DailyCited
How we label rumors
Unconfirmed
Circulating with at least one public source we can link to; no official confirmation.
Reported
Reported by a news outlet; not confirmed by the company.
Confirmed
Confirmed by the company or another official source.
Debunked
Denied by the company, retracted by the original source, or shown to be false by the evidence.

We record what sources show, with dates. We don't predict or score how likely a rumor is.

DailyCitedRSSAboutPrivacy