gamebench-2-results / benchmark-summary.json
DuelLab's picture
Publish gb2-20260713-16054c869f49
a182ce3 verified
Raw
History Blame Contribute Delete
990 Bytes
{
"schema_version": 2,
"public_release_id": "gb2-20260713-16054c869f49",
"generated_at": "2026-07-13T19:49:06Z",
"benchmark_line": "gamebench2",
"benchmark_version": "2-known8",
"scoring_policy_id": "entrant_identity_equal_mean_v7",
"rating_method_id": "sequential_elo",
"game_policy": "public_named",
"updated_at": "2026-07-13",
"model_count": 34,
"reasoning_variant_count": 80,
"game_count": 8,
"rated_match_count": 56163,
"top_reasoning_variants": [
{
"rank": 1,
"model_name": "Claude Fable 5",
"reasoning": "XHigh",
"score": 73.67927701948544
},
{
"rank": 2,
"model_name": "GPT-5.6 Sol",
"reasoning": "XHigh",
"score": 72.38492380259433
},
{
"rank": 3,
"model_name": "Claude Fable 5",
"reasoning": "Medium",
"score": 69.84775922173306
}
],
"data_license_url": "https://creativecommons.org/licenses/by/4.0/",
"citation_url": "https://duellab.org/citation"
}