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

metask-jev-4b

Jev rebuild · by Wayfind (metask-ai)

JevBench v1.4.2.2 score

47.8

Rank #12 of 91 ranked systems.

15.5 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
44.7
calibration
66.9
speed
89.1
cost
54.5

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, metask-jev-4b vs Jev 1.13.0. Intelligence: 44.7 vs 53.1; Calibration: 66.9 vs 76.3; Speed: 89.1 vs 83.3; Cost: 54.5 vs 52.0.50100Intelligence44.7 · 53.1Calibration66.9 · 76.3Speed89.1 · 83.3Cost54.5 · 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, metask-jev-4b vs Jev 1.13.0. Easy: 100% vs 100%; Standard: 98% vs 99%; Judge: 90% vs 95%; Hard: 59% vs 74%; Sealed: 28% vs 37%.50100Easy100% · 100%Standard98% · 99%Judge90% · 95%Hard59% · 74%Sealed28% · 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), metask-jev-4b vs Jev 1.13.0. Ambiguous / abstain: 24% vs 43%; Judge: 55% vs 54%; Long policy: 28% vs 44%; Multi-hop: 49% vs 64%; Probability: 44% vs 63%; Temporal / numeric: 21% vs 28%; Trade-off: 39% vs 55%; Routing: 100% vs 100%; Trap / adversarial: 80% vs 83%; Paraphrase: 14% vs 64%; Safety judge: 38% vs 38%.50100Ambiguous /abstain24% · 43%Judge55% · 54%Long policy28% · 44%Multi-hop49% · 64%Probability44% · 63%Temporal /numeric21% · 28%Trade-off39% · 55%Routing100% · 100%Trap /adversarial80% · 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, metask-jev-4b vs Jev 1.13.0. Ambiguous / abstain: 14% vs 30%; Judge: 39% vs 34%; Long policy: 28% vs 28%; Multi-hop: 32% vs 45%; Paraphrase: 14% vs 64%; Probability: 29% vs 50%; Safety judge: 38% vs 38%; Temporal / numeric: 21% vs 29%; Trade-off: 23% vs 38%; Trap / adversarial: 58% vs 42%.50100Ambiguous /abstain14% · 30%Judge39% · 34%Long policy28% · 28%Multi-hop32% · 45%Paraphrase14% · 64%Probability29% · 50%Safety judge38% · 38%Temporal /numeric21% · 29%Trade-off23% · 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 GPU (lium.io RTX 5090 32 GB); serial, in-process candidate-logit inference
Note on this row
Model card discloses 44.8k+16.1k+390 training rows incl. synthetic families intentionally mirroring JevBench hard families, plus repeated evaluation on all 231 public items (benchmark-directed development disclosed). Scan of 61 files: 0 exact matches.
Published source
https://github.com/metask-ai/metask-jev

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.