open-alternative-jev (Qwen3.5-4B, IkerMoel)v1.7 · 10 s rule
Data: JevBench v1.8.7, published 2026-11-03 · method v1.7 (10 s rule), re-scored from the recorded per-decision times
Capability 49.1 · rank #160 of 217 on the open-weights board
Last measured: 2026-10-02 · measurement release: v1.6.0
Apache-2.0 (code and weights)
intelligence
22.3
calibration
75.9
speed
90.9
cost
63.3
Composite (secondary)
9.4 · Composite rank #175
Cost and measurement conditions
$0.017 per 1,000 decisions (estimated)
v1.5.4 published cost per 1,000 decisions carried (pricing rules unchanged; v1.6 item lengths differ): ESTIMATE (proxy tokens, I-2): exact base-model market reference; reference deepinfra:Qwen/Qwen3.5-4B frozen at the 25 Sep cut-off; the snapshot records deprecation on 11 Jun 2026 and replacement by Qwen/Qwen3.5-9B; the frozen snapshot rates remain in use (I-3)
Measured p50 latency: 0.050 s (on-GPU, no network, loopback; not directly comparable with API latency, which we time end to end from our measuring host, network included)
Ranking and cap use: 0.25 s (est. incl. network (2x + 150 ms))
Self-hosted model: measured p50 0.050 s on-GPU, no network (timed on our GPU over loopback). Not directly comparable with API latency, which we time end to end from our measuring host, network included. The ranking and latency cap use the same estimate for every self-hosted row: 2 × measured + 150 ms = 0.25 s. Model authors can ask us to time a hosted endpoint instead (on request).
x2 + 0.15 s (assumption, not measured)
evaluator-owned Lium GPU pod (RTX5090), offline read-only container
Measured latency over 1,500 decisions: p50 0.050 s · p95 0.089 s
0 decisions over 10 s.
A decision that takes longer than 10 seconds (measured time) counts as a wrong answer and as zero calibration for that item: each type's calibration is multiplied by the share of its decisions answered within 10 s. The limit uses the measured time (raw). The displayed adjusted time for self-hosted rows (2 × + 0.15 s) is an assumption and is not used for the limit.
JevBench v1.6.43 values (method v1.6, before the 10 s rule)
Capability 49.14 · Composite 9.44 · rank #173 on the all-systems board · #134 open weights
intelligence 22.35 · calibration 75.94 · speed 91.30 · cost 63.28
v1.6 speed basis (124 items): p50 0.043 s · p95 0.082 s
These figures are kept for comparison. The current figures above are re-scored under the 10 s rule from the recorded per-decision times and replace them.