Bespoke Nimble 9B (Bespoke Labs)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 56.5 · rank #136 of 217 on the open-weights board
Last measured: 2026-10-02 · measurement release: v1.6.0
Apache-2.0 (weights); repository without a licence file as of 19 Sep
intelligence
43.1
calibration
69.9
speed
85.5
cost
56.4
Composite (secondary)
44.5 · Composite rank #104
Cost and measurement conditions
$0.028 per 1,000 decisions (estimated)
GPU-time estimate from v1.8.0: one monthly-rented RTX PRO 6000 96 GB, EUR 999/month / 730 hours, ECB USD 1.1133/EUR dated 28 Oct 2026, USD 1.5235/hour; average load 70%. Architecture: small qwen, 9B active parameters; throughput 27344.4444444 tokens/s (architecture-class calibration, 29 Oct 2026, efficient-engine prefill; capped at 100,000 tokens/s). The row's own scored cost basis: 2,079,172 input + 3,248 output tokens over 1,624 decisions = 1282.27832512 tokens/decision. GPU-time cost USD 0.0283501339224056/1,000; old list-price cost USD 0.128327832512315/1,000; applied min(old list-price cost, GPU-time cost) = USD 0.0283501339224056/1,000. The GPU-time estimate is lower. Estimated, not a bill. Intelligence, Calibration, measured latency and the common cost-item membership are unchanged. Prior prices remain in the archived v1.7 releases.
Measured p50 latency: 0.157 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: 0.46 s (est. incl. network (2x + 150 ms))
Self-hosted model: measured p50 0.157 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 = 0.46 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 (A6000), offline read-only container
Measured latency over 1,500 decisions: p50 0.157 s · p95 0.226 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 56.51 · Composite 21.08 · rank #124 on the all-systems board · #90 open weights
intelligence 43.14 · calibration 69.89 · speed 86.11 · cost 36.75
v1.6 speed basis (124 items): p50 0.145 s · p95 0.203 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.