Score · relative 0–100: 43.2
Rank: 48
Status: Partial 3/3
Players generated: 2026-07-02 – 2026-07-21
Match evidence updated: 2026-10-26
Results published: 2026-10-26

Compare in charts View player programs Back to leaderboard

Why this result is partial: one or more documented settings are untested or ineligible. The rank uses the available reasoning-setting results.

Scores by reasoning setting

High (Intensive group, tested and included)49.5
Medium (Balanced group, tested and included)38.6
Lownot included (Baseline group, tested, not included)41.5

Settings marked “not included” were tested but are left out of the overall score.

Available settings and grouping

Available settings

MinimalLowMediumHigh

Checked on 2026-07-21 · Provider documentation

How settings are grouped

Baseline: Minimal — available, not tested in the current results
Balanced: Medium — tested and included
Intensive: High — tested and included

Scores by game

Go43.8
Chess1 of 3 failed27.0
Shogi28.4
Xiangqi41.4
Hex61.8
International Draughts59.5
Game of the Amazons52.4
Tumbleweed37.2

Each game counts every generation attempt, a failed one as 0 (“1 of 2 failed” marks games with failed attempts). Scores are relative within each game, across the included settings; they do not compare game difficulty.

What this result shows

Strongest included setting: High (49.5).

Across included settings, the highest game score is Hex (61.8); the lowest is Chess (27.0). Each game counts every generation attempt, a failed one as 0. These are relative scores within each game, not direct comparisons of difficulty.

Match evidence measures these retained programs. It does not establish how a fresh generation will perform. Independent-generation variation is not estimated in this release.

Test protocol and evidence limits

Model identifier: gpt-5
Provider route (catalog): openai
Prompt identity / profile: Not retained in this result set / Not retained in this result set
Prompt version: Not retained in this result set
Generation policy: Generation-quality admission policy (see methodology)
Requested output-token limits: Not retained (0 of 19 indexed attempts)

Generation evidence

Medium: 7 contributing player programs across 7 games
High: 8 contributing player programs across 8 games

Counts describe contributing programs, not independent replications: repairs and resends are not new independent samples. Independent-generation lineage is not retained.

Game	Reasoning group	Programs	Working-program strength	Raw rating	Combined record (W / D / L)
Go	Intensive	1 working of 1 attempt	39.7	1370.8	152 / 86 / 278
Go	Balanced	1 working of 1 attempt	21.1	1166.6	69 / 65 / 382
Go	Baseline	1 working of 1 attempt	70.6	1678.5	325 / 57 / 134
Chess	Intensive	1 working of 1 attempt	42.3	1412.1	15 / 360 / 105
Chess	Baseline	1 working of 1 attempt	38.7	1383.8	5 / 346 / 129
Shogi	Intensive	1 working of 1 attempt	37.6	1319.6	91 / 154 / 249
Shogi	Balanced	1 working of 1 attempt	31.7	1266.4	45 / 191 / 258
Shogi	Baseline	1 working of 1 attempt	15.9	1090.3	32 / 72 / 390
Xiangqi	Intensive	1 working of 1 attempt	40.1	1397.9	26 / 331 / 141
Xiangqi	Balanced	1 working of 1 attempt	42.8	1419.5	71 / 268 / 159
Hex	Intensive	1 working of 1 attempt	75.9	1822.4	394 / 0 / 136
Hex	Balanced	1 working of 1 attempt	61.5	1585.9	317 / 0 / 213
Hex	Baseline	1 working of 1 attempt	47.8	1410.1	245 / 0 / 285
International Draughts	Intensive	1 working of 1 attempt	69.9	1751.6	332 / 7 / 163
International Draughts	Balanced	1 working of 1 attempt	49.2	1406.5	216 / 17 / 269
Game of the Amazons	Intensive	1 working of 1 attempt	52.9	1484.4	268 / 0 / 262
Game of the Amazons	Balanced	1 working of 1 attempt	62.8	1614.2	323 / 0 / 207
Game of the Amazons	Baseline	1 working of 1 attempt	41.5	1341.2	206 / 0 / 324
Tumbleweed	Intensive	1 working of 1 attempt	37.2	1247.7	76 / 218 / 244
Tumbleweed	Balanced	1 working of 1 attempt	39.8	1276.2	143 / 111 / 284
Tumbleweed	Baseline	1 working of 1 attempt	34.7	1221.3	111 / 123 / 304

Each row covers all the model's programs in one game and setting. Strength and raw rating are attempt-weighted means; under them, each program's score and the lowest and highest rating describe the programs (not a confidence interval). The combined record adds up their matches: evidence volume, not strength.

Failures and budgets

1 model-code generation failures; 0 provider/infrastructure failures excluded from the generation denominator. Runtime player faults are separate: 1 recorded one-sided faults across contributing program rows. A crash or illegal move pinned on one program also counts as that program's loss in its match record; other faults, such as timeouts, are left out of rating evidence.

Repair/resend outcomes and available costs are included. Exact retry ceilings and execution budgets are not retained in this result set; see the methodology for the public policy. Do not infer a measured setting from a current provider catalog.

Player program results

High: 75% / 25% / 0% · n 8Medium: 75% / 13% / 12% · n 8Low: 67% / 33% / 0% · n 6 All three benchmark settings tested

Generation cost

$0.19average estimated cost per game and reasoning setting

Median: $0.17
Range: $0.07–$0.45
Known cost subtotal (incomplete or unverified charges): $4.46
Cost data: 24 combinations · 24 runs
Price source: Standard list-price estimate
Price list: boardgame-list-prices-2026-07-12 + boardgame-list-prices-2026-10-09
Average output: 22k tokens
Output range: 13k–44k tokens
Runs with output-token data: 16

How cost is measured

Comparison cost gives equal weight to each model, game, and reasoning-setting combination. Recorded generation attempts include repairs and resends when cost evidence is available; runtime compute and service operation are excluded. Cost and output-token counts use separate telemetry denominators. A list-price estimate is not actual expenditure; mixed evidence is not a uniform standard-price comparison.

Test protocol

Combines this model's tested reasoning settings
Player-program generation; retained protocol details are under “Test protocol and evidence limits” above

GPT-5