Chon-1.0-E4B
Data: JevBench v1.6.40, published 2026-10-22
Capability 68.4 · Not ranked in the current release
latency 72.2× Jev (v1.5 reference)
Last measured: 2026-10-19 · measurement release: v1.6.32
Apache-2.0 (adapter card and base licence agree)
intelligence
55.0
calibration
81.8
speed
44.0
cost
70.7
Composite (secondary)
45.9 · Composite rank #43
Cost and measurement conditions
$0.0094 per 1,000 decisions (not published)
Exact listed rate of the base model google/gemma-4-E4B-it (DeepInfra, frozen 25 Sep 2026 catalog snapshot, USD 0.02/M input, 0.10/M output); this line is added to the frozen pricing module at its next re-freeze. No output tokens under the author's --no-think mode. COMMON COST BASIS: the published cost uses the same remaining 1,485 items as every other row: 701,161 input tokens, USD 0.009443245791245792 per 1,000 decisions (0.146x cap). ALL-ITEMS DIAGNOSTIC: USD 0.010571213333333333 per 1,000 over all 1,500 items (0.164x cap). All items still count for Intelligence.
Measured p50 latency: 22.177 s (on-CPU, 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: 44.50 s (est. incl. network (2x + 150 ms))
Self-hosted model: measured p50 22.177 s on-CPU, no network (timed on our CPU host 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 = 44.50 s. Model authors can ask us to time a hosted endpoint instead (on request).
x2 + 0.15 s (assumption, not measured)
Our own CPU server (AMD EPYC, 32 cores), disposable container with no network; the network cut was proven inside the container before it read any sealed item. Served by the author's own chon.serve 1.0.0 in its documented CPU mode (--device cpu --no-think). Published Speed applies the standard self-hosted x2 + 0.15 s adjustment (assumption, not measured).