JevBench public 231 accuracy and latency (Mac, 2026-09-24)

Same subsection as coreai-model-zoo a708c2e (models/openthai-systemone/README.md).

README.md

@@ -113,6 +113,23 @@ No other Core AI, Python or Swift engine job appeared in the before/after proces
 benchmark needed a local CLI option to select `EngineOptions.variant`; the trial loop was
 unchanged. [Trials, load times and environment](https://github.com/john-rocky/coreai-model-zoo/blob/main/models/openthai-systemone/llm-benchmark.json).
 
+### JevBench public 231 (Mac, 2026-09-24)
+
+| easy 48 | standard 72 | hard 111 | ECE hard | p50 | p95 | hard max |
+|---:|---:|---:|---:|---:|---:|---:|
+| 1.000 | 0.819 | 0.324 | 0.449 | 0.62 s | 25.93 s | 39.0 s |
+
+The benchmark's own harness
+([fstandhartinger/jevbench](https://github.com/fstandhartinger/jevbench) `2fa63fa`, v1.4.0,
+`typesafe` adapter) ran the 231 public items against coreai-kit `adbc755` `decide-cli serve`
+with the ship bundle, one question per request. Accuracy per tier and the hard tier's ECE are
+JevBench's own scoring (argmax of the returned probabilities); p50 and p95 are per-request
+latency over all 231 requests, hard max the maximum over the hard tier. Latency was measured
+without an exclusive GPU window (contended), with this model's server running alone; the
+graph's prefill is S=1, and at that kit commit the state was prefilled again for every
+question. JevBench's published scores (Intelligence and the rest) are chance-corrected over 534
+items, sealed ones included, and are not comparable to these accuracies.
+
 ## Through the kit
 
 **Measured through coreai-kit**, using its sequential engine and tokenizer: `decide-cli parity`