Not evaluated
Prime number theorem (lower bound on the score functional): 0.99490
Previous 0.92129, AlphaEvolve
The gist
For the lower bound on the score functional linked to the prime number theorem, Together AI and EinsteinArena agents give 0.99490, against 0.92129 for AlphaEvolve (June 2025); larger is better; no DiscoverIA status yet.
DiscoverIA status on October 12, 2026: not evaluated, meaning no third-party evaluation found.
AI draft, not reviewedWhat has not been reviewed?
Text written by an AI (DiscoverIA pipeline), on . The status and its sources come from the registry; only the explanatory text has not been reviewed. A reported error is corrected, then the text is reviewed.
Cite this entry
A single status snapshot since Oct 12, 2026: a citation always points to a dated snapshot.
Cite snapshot 1, Oct 12, 2026
APA citation with the dated link
BibTeX @misc, key discoveria-ea-prime-number-theorem-r1
@misc{discoveria-ea-prime-number-theorem-r1,
author = {{DiscoverIA}},
title = {Prime number theorem (lower bound on the score functional): 0.99490 (previous 0.92129, AlphaEvolve)},
howpublished = {DiscoverIA registry, entry ea-prime-number-theorem},
note = {Status “Not evaluated”, snapshot 1, recorded on October 12, 2026. New result},
year = {2026},
url = {https://discoveria.ai/entry/ea-prime-number-theorem#releve-1},
urldate = {2026-10-12}
}Dated link opens this snapshot in the history
To understand ≈ 20 s reading time, estimated
What it changes
If the value holds, AlphaEvolve's score (0.92129, June 2025) would be clearly exceeded, at 0.99490. The repository already gave 0.994179 in March 2026, with 1 785 keys.
What remains unclear
The paper says it chose this problem because the best known human solutions are much better than those of AI: the previous record compared against is AlphaEvolve's, not the best human result. The verifier is a public program, not a third party. The online ranking shows 0.9976572852677297 (CodexProLong, August 15, 2026), different from the paper's value.
Read the paper
Harnessing the Collective Intelligence of AI Agents in the Wild for New Discoveries
What it establishes : The authors present EinsteinArena and report that by May 2026 its agents had set 11 new records across 15 open problems, including the kissing number in dimension 11, raised from 593 to 604.
Where to start : The abstract, then Table 1 (Section 2.2): the 15 problems, with the previous and new records. Section 3 and Appendix B cover dimension 11 in detail, Appendix A each problem. The arXiv abstract page announces 12 records, the paper itself 11.
Written by an AI · reviewed by a human
For experts
Sources, method, key facts, status history, report an error
Status
Not evaluated
Recorded on · arxiv.org
Step 1 of 4 on the scale: not evaluated. Next step: under review.
- Announced
- Recordednot evaluated
- nextTo follow
What this status means
Source reading : Source read word for word (level 3 of 3)
The result is declared neither right nor wrong.
To move to “Under review”: a public review under way. To “Peer reviewed”: a journal publication or a review by named experts.
Sources
- Main sourcearxiv.orgpage read word for word · consulted on Oct 12, 2026 ·
we additionally report the previous best known result and the current best result
- Papertogether.aipage read word for word · consulted on Oct 12, 2026 ·
we currently have the best known results for the following problems
- Repositorygithub.comlink only, reading not recorded
- Repositoryeinsteinarena.com · api/problems/prime-number-theoremlink only, reading not recorded
- Repositoryeinsteinarena.com · bestlink only, reading not recorded
- Repositoryeinsteinarena.com · problems/prime-number-theoremlink only, reading not recorded
Method
DiscoverIA status: not evaluated, meaning that no third-party evaluation has been found in the sources read. Set on October 12, 2026 from arxiv.org (page read, quotation checked word for word: “we additionally report the previous best known result and the current best result”).
Limits
No named third party: the recorded checks come from the claimant or from a platform tool. DiscoverIA has not checked this result itself.
Key facts
- Claimed by
- Together AI and EinsteinArena agents (agent not named by the source for the paper's value)
announced on Mar 2026 - Nature
- new result
- Field
- Mathematics (analytic number theory / partial functions and a constant tied to pi(x)·log x / x)
- Checked by
- the platform's public checker (a tool); no third party named
- Sources
- 6 : 2 read word for word, 4 links only
- This entry
- AI draft, not reviewed
Status history
Only one observation so far.
- 01
First observation:Not evaluatedCurrent status
not evaluated: only the platform's public checker, no third party named
Source: arxiv.orgread word for word
An observation is added; none is ever erased.
In the same set
EinsteinArena: a platform where AI agents tackle open problems with public verifiers · 16 entries
- Not evaluated : Third autocorrelation inequality: upper bound 1.4523 (previous 1.4556, AlphaEvolve)
- Not evaluated : Flat polynomials of degree 69: 1.28093 (previous 1.34093, AlphaEvolve)
- Not evaluated : Distance ratio of 16 points in the plane: 12.889230 (previous 12.889266, AlphaEvolve)
- Not evaluated : Prime number theorem (lower bound on the score functional): 0.99490 (previous 0.92129, AlphaEvolve)
- Not evaluated : Circles in a rectangle, 21 of them: sum of radii 2.365832255 (previous 2.365832133, AlphaEvolve)
- Not evaluated : Tammes problem (50 points on the sphere): 0.5134721 (previous 0.5134719, AlphaEvolve)
- Not evaluated : Edges versus triangles (minimum triangle density): -0.71171 (previous -0.71249, AlphaEvolve)
Related frontiers
Report an error or suggest a source
Thank you, your message was sent.
Sending failed. Try again, or write to contact@discoveria.ai.
Is your result listed?What each status requires, correcting this entry, replying to it.