Completed / CodeSOTA execution on White
Inspect the measurement.
OmniJev/PlayJev-0.8B · ARC-Easy / Qwen3.5 same-base cohort · 27.6% accuracy
Full benchmark split · 2376 scored examples. Full raw-accuracy text-only diagnostic for a same-base Qwen3.5 cohort. Stock trusted model classes, no chat or images; generic text scores do not validate specialist OCR/classification/retrieval uses. Unique total/text/vision parameters measured after loading; source-reported metrics are not imported.
Run ID / completed
0f4e12b0cba7a4a4a9b3603c
17 Oct 2026, 03:45 UTC
17 Oct 2026, 03:45 UTC
Model revision
a7348002b1e159add7037d6d50812cd4db2d96e9Comparison scope
84fe51c6790b19376ad4a234c742b80e0f73e3eb0d6a7257d48e5315d6a7c078Evaluator SHA-256
18c73508df97c7b2334e8d47512c9f06edeccf9d2119d42826f8648a7b114303Observed resources
91s elapsed · 3.90 GiB peak torch allocation
Dollar cost: not measured
Dollar cost: not measured
Download evidence
Inspect individual examples
Read the recorded prediction, reference and outcome for each example.
- Scored examples
- 2,376
- Passed
- 655
- Failed
- 1,721
- Metric
- accuracy
Examples are downloaded only when requested. Search covers all loaded records; 25 examples are displayed per page. Download all examples · JSONL
Example evidence has not been loaded.