Ethereum’s better.codes contest now measures a cryptographic proof gap that researchers can attack from both sides, bringing the community closer to resolving a critical challenge in zero-knowledge proof technology.
At 15:44:47 UTC on Aug. 21, the live leaderboard displayed a 63.99-bit lower certificate and a 116.13-bit upper certificate for the koalaIRS12 benchmark. These competing metrics left a 52.14-bit interval unresolved after nine promoted submissions from seven independent solvers, highlighting the ongoing collaborative effort to narrow the gap.
KoalaIRS12 represents a fixed parameter profile for an interleaved Reed–Solomon reduction used in proof-system research. While the challenge repository defines its score as a spot-check quantity, it explicitly excludes interpreting this score as a direct measure of whole-system soundness or full-protocol security. Instead, it serves as a public, reproducible metric showing the precise distance between what the challenge has proved safe and what its upper certificate still rules unsafe.
How the Leaderboard Closes the Interval
The contest utilizes two distinct tracks to systematically close the cryptographic interval:
- The Soundness Track: This track works to raise the lower certificate. At a certified radius, a successful submission proves that the benchmark’s executable reduction-error bound meets the encoded target, subsequently mapping that radius to the score displayed on the board.
- The Attack Track: This track aims to lower the upper certificate. Its theorem certifies an unsafe suffix under the benchmark’s winning-set-density condition. The repository covers that suffix directly because the formalization assumes no monotonicity theorem for winning-set density.
Machine-checked results ensure that the certified scope remains rigorous. While the upper-certificate track’s score describes the formal boundary for koalaIRS12, calculating an actual Ethereum attack cost would require a separate, comprehensive whole-system analysis.
The Ethereum Foundation’s launch announcement emphasizes that the theorem statement, parameter point, and verification harness are pinned. Each submission exports the required theorem, where a comparator checks the statement against the target, and the Lean kernel verifies the proof before promotion. While an accepted result proves the submitted theorem inside this pinned environment, transitioning to production assurance requires covering the model’s completeness, the assumptions embedded in its definitions, implementation fidelity, and the composition of separately analyzed components.
The Path to Ethereum’s December Target
The Foundation’s December 2025 zkEVM security roadmap called for 128-bit provable security, a final proof size of 300 KiB or less, and a formal soundness argument for the recursion architecture. Following a February security-sprint update, the M3 deadline was moved to early December 2026, aligning the architecture-security argument with a Dec. 1 deliverable.
For the koalaIRS12 benchmark, a lower certificate reaching the encoded 128-bit target would settle the soundness side of this benchmark at its fixed parameter point. However, a production zkEVM claim would additionally require soundness accounting across every relevant component, proof-size compliance, a documented recursion topology, an argument for how its parts compose, and evidence that specifications match implementations.
The Foundation’s public progress page, last synced on Aug. 20, lists zkVM readiness and ISA compliance results, naming real-time proving and soundcalc integration as key criteria. Currently, the 52.14-bit interval stands as a live measure of unfinished work on koalaIRS12, and Ethereum’s ultimate 128-bit production case will depend on how this component-level evidence fits into the larger proof.
Academic research, such as the paper by Gal Arnon, Dan Boneh, and Giacomo Fenzi, continues to explore foundational questions surrounding list decoding, Reed–Solomon proximity gaps, correlated agreement, and mutual correlated agreement. These open questions remain critical to the broader succinct proof systems landscape, independent of specific leaderboard scores.
On better.codes, soundness submissions can continue to lift the 63.99-bit lower certificate, while attack submissions can pull the 116.13-bit upper certificate down. Across the broader zkEVM roadmap, teams are tasked with publishing the system-level accounting, proof sizes, recursion arguments, and implementation evidence required for the early-December review. This two-sided movement ensures that the interval remains a dynamic, informative metric for the community as it races toward the December milestone.
