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

JevK5 v0.2.0

Jev rebuild · by allebee

JevBench v1.4.2.2 score

62.0

Rank #5 of 91 ranked systems.

1.2 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
48.9
calibration
74.5
speed
91.1
cost
59.5

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.

The four score axes

Radar: the four score axes, two systemsThe four score axes, JevK5 v0.2.0 vs Jev 1.13.0. Intelligence: 48.9 vs 53.1; Calibration: 74.5 vs 76.3; Speed: 91.1 vs 83.3; Cost: 59.5 vs 52.0.50100Intelligence48.9 · 53.1Calibration74.5 · 76.3Speed91.1 · 83.3Cost59.5 · 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, JevK5 v0.2.0 vs Jev 1.13.0. Easy: 100% vs 100%; Standard: 96% vs 99%; Judge: 95% vs 95%; Hard: 70% vs 74%; Sealed: 33% vs 37%.50100Easy100% · 100%Standard96% · 99%Judge95% · 95%Hard70% · 74%Sealed33% · 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), JevK5 v0.2.0 vs Jev 1.13.0. Ambiguous / abstain: 53% vs 43%; Judge: 57% vs 54%; Long policy: 36% vs 44%; Multi-hop: 58% vs 64%; Probability: 48% vs 63%; Temporal / numeric: 30% vs 28%; Trade-off: 42% vs 55%; Routing: 100% vs 100%; Trap / adversarial: 85% vs 83%; Paraphrase: 14% vs 64%; Safety judge: 38% vs 38%.50100Ambiguous /abstain53% · 43%Judge57% · 54%Long policy36% · 44%Multi-hop58% · 64%Probability48% · 63%Temporal /numeric30% · 28%Trade-off42% · 55%Routing100% · 100%Trap /adversarial85% · 83%Paraphrase14% · 64%Safety judge38% · 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, JevK5 v0.2.0 vs Jev 1.13.0. Ambiguous / abstain: 43% vs 30%; Judge: 37% vs 34%; Long policy: 23% vs 28%; Multi-hop: 39% vs 45%; Paraphrase: 14% vs 64%; Probability: 39% vs 50%; Safety judge: 38% vs 38%; Temporal / numeric: 27% vs 29%; Trade-off: 23% vs 38%; Trap / adversarial: 58% vs 42%.50100Ambiguous /abstain43% · 30%Judge37% · 34%Long policy23% · 28%Multi-hop39% · 45%Paraphrase14% · 64%Probability39% · 50%Safety judge38% · 38%Temporal /numeric27% · 29%Trade-off23% · 38%Trap /adversarial58% · 42%
Share correct within each sealed family — system-level aggregates; the items stay private.

Availability and evidence

Openness
Unknown in the published row
License note
Apache-2.0 (code/adapter); Apache-2.0 (Qwen3.5-4B base)
Cost evidence
estimated; the board’s row disclosure contains the published basis.
Endpoint condition
re-run on a throwaway RunPod pod with the original recipe; deviations in its manifest
Note on this row
Author says no JevBench items or outputs were used for training, tuning, or selection; public results are reported. Scan found only one generic instruction shared by 8 public hard items; unreleased teacher/replay corpora were unavailable.
Published source
https://github.com/allebee/jevk5

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.