Benchmarks
More
OverviewBenchmaxxingCompareChartsCost vs CapabilityProviders per ModelProvider explorerGatewaysEU & SovereignJev-class modelsAbout
JevBench by Benchmark Heaven · v1.4.2 · individual system

Kushal Patil — Gemma 4 31B IT (Autoloops)

This detail uses the public, hash-checked v1.4.2 aggregate. Scores and ranks can change when a new release is published; the page preview remains name-only.

JevBench v1.4.2 score

30.0

Rank #42 of 89 ranked systems.

34.2 points behind Jev 1.13.0's 64.1.

Where it sits among the 89 ranked systems. The marked tick is Jev 1.13.0 (63.3).

Published axes

intelligence
59.8
calibration
79.2
speed
84.0
cost
36.0

Accuracy per tier, incl. sealed

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: Kushal Patil — Gemma 4 31B IT — Jev (TypeSafe, closed) · Score 30.0 (#42)
  • B: decider-4b v2 — system-one-open · Score 64.1 (#1)

The four score axes

Radar: the four score axes, two systemsThe four score axes, Kushal Patil — Gemma 4 31B IT vs decider-4b v2. Intelligence: 59.8 vs 49.4; Calibration: 79.2 vs 75.0; Speed: 84.0 vs 92.9; Cost: 36.0 vs 60.9.50100Intelligence59.8 · 49.4Calibration79.2 · 75.0Speed84.0 · 92.9Cost36.0 · 60.9
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, Kushal Patil — Gemma 4 31B IT vs decider-4b v2. Easy: 100% vs 100%; Standard: 99% vs 97%; Judge: 96% vs 88%; Hard: 83% vs 67%; Sealed: 46% vs 35%.50100Easy100% · 100%Standard99% · 97%Judge96% · 88%Hard83% · 67%Sealed46% · 35%
Share correct per tier; Sealed = the 308 private decisions, aggregate only.

Hard tier by family (v1.2 topics)

Kushal Patil — Gemma 4 31B IT and decider-4b v2 have no published v1.2 hard-tier family breakdown.

Neither selected system has a published series for this view.

Share correct within each family of the 47 v1.2 hard-tier decisions (public and held-out).

Sealed set by family

Radar: sealed set by family, two systemsSealed set by family, Kushal Patil — Gemma 4 31B IT vs decider-4b v2. Ambiguous / abstain: 46% vs 43%; Judge: 73% vs 32%; Long policy: 33% vs 30%; Multi-hop: 53% vs 34%; Paraphrase: 57% vs 50%; Probability: 46% vs 36%; Safety judge: 44% vs 38%; Temporal / numeric: 29% vs 30%; Trade-off: 42% vs 35%; Trap / adversarial: 50% vs 33%.50100Ambiguous /abstain46% · 43%Judge73% · 32%Long policy33% · 30%Multi-hop53% · 34%Paraphrase57% · 50%Probability46% · 36%Safety judge44% · 38%Temporal /numeric29% · 30%Trade-off42% · 35%Trap /adversarial50% · 33%
Share correct within each sealed family — system-level aggregates; the items stay private.

Availability and evidence

Openness
Marked closed in the published row
License note
Gemma terms of use (Google) for the base weights; hosted Autoloops service
Cost evidence
measured; the board’s row disclosure contains the published basis.
Endpoint condition
Autoloops production API, reached from a Hetzner server in Germany

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