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

lev-350m (Franck Verrot, LFM2.5-350M)

Jev rebuild · by Franck Verrot (franckverrot)

JevBench v1.4.2.2 score

28.5

Rank #46 of 91 ranked systems.

34.8 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
34.8
calibration
70.6
speed
85.3
cost
76.1

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: lev-350m — Jev rebuild · Score 28.5 (#46)
  • 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, lev-350m vs Jev 1.13.0. Intelligence: 34.8 vs 53.1; Calibration: 70.6 vs 76.3; Speed: 85.3 vs 83.3; Cost: 76.1 vs 52.0.50100Intelligence34.8 · 53.1Calibration70.6 · 76.3Speed85.3 · 83.3Cost76.1 · 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, lev-350m vs Jev 1.13.0. Easy: 99% vs 100%; Standard: 72% vs 99%; Judge: 69% vs 95%; Hard: 37% vs 74%; Sealed: 25% vs 37%.50100Easy99% · 100%Standard72% · 99%Judge69% · 95%Hard37% · 74%Sealed25% · 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), lev-350m vs Jev 1.13.0. Ambiguous / abstain: 20% vs 43%; Judge: 41% vs 54%; Long policy: 21% vs 44%; Multi-hop: 21% vs 64%; Probability: 27% vs 63%; Temporal / numeric: 23% vs 28%; Trade-off: 42% vs 55%; Routing: 60% vs 100%; Trap / adversarial: 55% vs 83%; Paraphrase: 43% vs 64%; Safety judge: 25% vs 38%.50100Ambiguous /abstain20% · 43%Judge41% · 54%Long policy21% · 44%Multi-hop21% · 64%Probability27% · 63%Temporal /numeric23% · 28%Trade-off42% · 55%Routing60% · 100%Trap /adversarial55% · 83%Paraphrase43% · 64%Safety judge25% · 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, lev-350m vs Jev 1.13.0. Ambiguous / abstain: 16% vs 30%; Judge: 32% vs 34%; Long policy: 20% vs 28%; Multi-hop: 13% vs 45%; Paraphrase: 43% vs 64%; Probability: 21% vs 50%; Safety judge: 25% vs 38%; Temporal / numeric: 23% vs 29%; Trade-off: 42% vs 38%; Trap / adversarial: 42% vs 42%.50100Ambiguous /abstain16% · 30%Judge32% · 34%Long policy20% · 28%Multi-hop13% · 45%Paraphrase43% · 64%Probability21% · 50%Safety judge25% · 38%Temporal /numeric23% · 29%Trade-off42% · 38%Trap /adversarial42% · 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 (code); weights under LiquidAI's LFM1.0 licence, following the LFM2.5-350M base
Cost evidence
estimated; the board’s row disclosure contains the published basis.
Endpoint condition
our RunPod GPU (L40 48 GB, Czechia), reached over the internet from Germany
Note on this row
Weights franckverrot/lev-350m revision ab08ad8b8f346994d983152917e114224f6adac7, code github.com/franckverrot/lev c48a945dbf629998d7458dcc5c16f58df964db94, the author's own lev.serve /v1/systemone endpoint with its shipped calibration temperature, base LiquidAI/LFM2.5-350M, on our RunPod L40 in Czechia Offline local open-weight inference on the frozen 308-item v1.4 set in a network-disabled, read-only container; no operator endpoint received sealed text.
Published source
https://github.com/franckverrot/lev

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.