Benchmarks
More
OverviewBenchmaxxingCompareChartsCost vs CapabilityProviders per ModelProvider explorerGatewaysEU & SovereignJev-class modelsImage JevBenchAbout
JevBench by Benchmark Heaven · v1.4.2.2 · individual system

Laya (Convai Innovations, ModernBERT-large 421M)

Jev rebuild · by Convai Innovations

JevBench v1.4.2.2 score

30.3

Rank #43 of 91 ranked systems.

33.0 points behind Jev 1.13.0's 63.3.

Where it sits among the 91 ranked systems. The marked tick is Jev 1.13.0 (63.3).

Published axes

intelligence
36.1
calibration
63.7
speed
71.1
cost
86.2

Against Jev 1.13.0

Four radars compare this fixed pair across the score axes, accuracy per tier, and accuracy by family on the hard tier and sealed set. Further out is better on every spoke.

  • A: Laya — Jev rebuild · Score 30.3 (#43)
  • B: Jev 1.13.0 — Jev (TypeSafe, closed) · Score 63.3 (#4)

The four score axes

Radar: the four score axes, two systemsThe four score axes, Laya vs Jev 1.13.0. Intelligence: 36.1 vs 53.1; Calibration: 63.7 vs 76.3; Speed: 71.1 vs 83.3; Cost: 86.2 vs 52.0.50100Intelligence36.1 · 53.1Calibration63.7 · 76.3Speed71.1 · 83.3Cost86.2 · 52.0
0–100, the values in the table. A label-only system has no calibration (counted as 0).

Accuracy per tier, incl. sealed

Radar: accuracy per tier, incl. sealed, two systemsAccuracy per tier, incl. sealed, Laya vs Jev 1.13.0. Easy: 94% vs 100%; Standard: 73% vs 99%; Judge: 69% vs 95%; Hard: 34% vs 74%; Sealed: 31% vs 37%.50100Easy94% · 100%Standard73% · 99%Judge69% · 95%Hard34% · 74%Sealed31% · 37%
Share correct per tier; Sealed = the 308 private decisions, aggregate only.

Current question set by family (hard + sealed)

Radar: current question set by family (hard + sealed), two systemsCurrent question set by family (hard + sealed), Laya vs Jev 1.13.0. Ambiguous / abstain: 25% vs 43%; Judge: 38% vs 54%; Long policy: 29% vs 44%; Multi-hop: 32% vs 64%; Probability: 29% vs 63%; Temporal / numeric: 23% vs 28%; Trade-off: 42% vs 55%; Routing: 30% vs 100%; Trap / adversarial: 45% vs 83%; Paraphrase: 50% vs 64%; Safety judge: 31% vs 38%.50100Ambiguous /abstain25% · 43%Judge38% · 54%Long policy29% · 44%Multi-hop32% · 64%Probability29% · 63%Temporal /numeric23% · 28%Trade-off42% · 55%Routing30% · 100%Trap /adversarial45% · 83%Paraphrase50% · 64%Safety judge31% · 38%
Share correct per family across the 220 hard-tier decisions (public and held out) and the 308 sealed decisions of v1.4, pooled; Routing is hard-tier only, Paraphrase and Safety judge sealed only.

Sealed set by family

Radar: sealed set by family, two systemsSealed set by family, Laya vs Jev 1.13.0. Ambiguous / abstain: 30% vs 30%; Judge: 39% vs 34%; Long policy: 28% vs 28%; Multi-hop: 29% vs 45%; Paraphrase: 50% vs 64%; Probability: 21% vs 50%; Safety judge: 31% vs 38%; Temporal / numeric: 20% vs 29%; Trade-off: 38% vs 38%; Trap / adversarial: 58% vs 42%.50100Ambiguous /abstain30% · 30%Judge39% · 34%Long policy28% · 28%Multi-hop29% · 45%Paraphrase50% · 64%Probability21% · 50%Safety judge31% · 38%Temporal /numeric20% · 29%Trade-off38% · 38%Trap /adversarial58% · 42%
Share correct within each sealed family — system-level aggregates; the items stay private.

Availability and evidence

Openness
Code and weights marked open in the published row
License note
Apache-2.0
Cost evidence
estimated; the board’s row disclosure contains the published basis.
Endpoint condition
our CPU (4 threads, Ryzen 5 3600)
Note on this row
The English checkpoint (repo root), run on our CPU through its own `laya` package. Its budget is 512 tokens per question, so long hard-tier states are cut by the package itself.
Published source
https://huggingface.co/convaiinnovations/laya

From the public v1.4.2.2 aggregate. Scores and ranks can change when a new release is published.

Read the full board and published method. The overall score is a composite, not raw accuracy.