Score · relative 0–100: 81.1
Rank: 7
Status: Complete 3/3
Rank change: New
Players generated: 2026-11-01
Match evidence updated: 2026-11-03

Compare in charts View player programs Back to leaderboard

XHighIntensive
85.7 · rank 7 on the Intensive leaderboard
Cost per program
$0.107 · list-price estimate
Working programs
8 of 8
API time
Not recorded
7.5 points above Medium, at 2.7× the cost
- Higher score for $0.107 or less: Pareto 26.10 (no Overall rank) Provider default (reasoning controls unavailable) · Baseline: 89.7 for $0.0296 (provider-reported)
- It is also the cheapest of the 3 offered settings scoring at least as high for less.
MediumBalanced
78.2 · rank 6 on the Balanced leaderboard
Cost per program
$0.0395 · list-price estimate
Working programs
8 of 8
API time
Not recorded
1.2 points below None, at 1.5× the cost
- Higher score for $0.0395 or less: Pareto 26.10 (no Overall rank) Provider default (reasoning controls unavailable) · Baseline: 89.7 for $0.0296 (provider-reported)
- It is also the cheapest of the 2 offered settings scoring at least as high for less.
NoneBaseline
79.4 · rank 6 on the Baseline leaderboard
Cost per program
$0.0270 · list-price estimate
Working programs
8 of 8
API time
Not recorded
On the best-value line across ranked settings.
- No ranked setting scores higher for $0.0270 or less.
- No ranked setting scores at least as high for less.

Score and cost of every ranked setting

The chart needs JavaScript. The setting cards hold this model's results, and Charts compares every model.

Each setting is placed by its published score and its generation cost per player program, the mean over the game cells including repairs and failed attempts. Higher is better; further left costs less (logarithmic axis, with recorded $0 in its own lane). Filled marks are compared; hollow marks are not (the reason is in the table). The line marks the best value across ranked reasoning settings: no ranked setting with complete cost evidence scores at least as high for the same or less. It includes models without an Overall rank, so it can differ from the line on Charts. Costs mix provider-reported charges and list-price estimates; differences in score points are not tested for significance.

Show the numbers

The table needs JavaScript.

Available settings and grouping

Available settings

NoneLowMediumHighXHighMax

Checked on 2026-10-31 · Provider documentation

How settings are grouped

Baseline: None — tested and included
Balanced: Medium — tested and included
Intensive: XHigh — tested and included

Scores by game

Go65.2
Chess72.6
Shogi74.9
Xiangqi81.6
Hex91.0
International Draughts89.0
Game of the Amazons94.6
Tumbleweed80.0

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: XHigh (85.7).

Across included settings, the highest game score is Game of the Amazons (94.6); the lowest is Go (65.2). 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-6.1-terra
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: 128000 (24 of 24 indexed attempts)

Generation evidence

None: 8 contributing player programs across 8 games
Medium: 8 contributing player programs across 8 games
XHigh: 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	73.4	1733.8	365 / 38 / 121
Go	Balanced	1 working of 1 attempt	61.3	1598.7	281 / 80 / 165
Go	Baseline	1 working of 1 attempt	60.7	1593.0	271 / 95 / 160
Chess	Intensive	1 working of 1 attempt	76.8	1721.3	282 / 183 / 27
Chess	Balanced	1 working of 1 attempt	70.5	1652.1	233 / 214 / 43
Chess	Baseline	1 working of 1 attempt	70.6	1653.7	221 / 245 / 28
Shogi	Intensive	1 working of 1 attempt	88.3	1975.1	420 / 59 / 33
Shogi	Balanced	1 working of 1 attempt	59.5	1530.6	201 / 183 / 130
Shogi	Baseline	1 working of 1 attempt	76.8	1738.2	338 / 89 / 83
Xiangqi	Intensive	1 working of 1 attempt	83.6	1814.9	404 / 45 / 65
Xiangqi	Balanced	1 working of 1 attempt	78.3	1741.4	351 / 90 / 71
Xiangqi	Baseline	1 working of 1 attempt	83.0	1804.6	396 / 50 / 66
Hex	Intensive	1 working of 1 attempt	95.1	2546.7	519 / 0 / 25
Hex	Balanced	1 working of 1 attempt	90.1	2303.0	493 / 0 / 53
Hex	Baseline	1 working of 1 attempt	87.9	2217.2	475 / 0 / 65
International Draughts	Intensive	1 working of 1 attempt	89.9	2225.1	454 / 18 / 42
International Draughts	Balanced	1 working of 1 attempt	90.6	2244.2	452 / 33 / 31
International Draughts	Baseline	1 working of 1 attempt	86.4	2134.6	425 / 35 / 52
Game of the Amazons	Intensive	1 working of 1 attempt	97.0	2622.8	525 / 0 / 13
Game of the Amazons	Balanced	1 working of 1 attempt	95.6	2527.3	520 / 0 / 20
Game of the Amazons	Baseline	1 working of 1 attempt	91.3	2287.7	496 / 0 / 44
Tumbleweed	Intensive	1 working of 1 attempt	81.4	1998.8	439 / 0 / 103
Tumbleweed	Balanced	1 working of 1 attempt	80.0	1967.2	430 / 0 / 110
Tumbleweed	Baseline	1 working of 1 attempt	78.6	1937.4	424 / 0 / 118

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

0 model-code generation failures; 0 provider/infrastructure failures excluded from the generation denominator. Runtime player faults are separate: 0 recorded one-sided faults across contributing program rows. A crash, an illegal move, or a timeout repeated by a later quiet replay, pinned on one program, also counts as that program's loss in its match record; other faults 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

XHigh: 100% / 0% / 0% · n 8Medium: 100% / 0% / 0% · n 8None: 100% / 0% / 0% · n 8 All three benchmark settings tested

Generation cost, all settings combined

$0.06average estimated cost per game and reasoning setting

Median: $0.04
Range: $0.02–$0.13
Estimated total for recorded runs: $1.39
Cost data: 24 combinations · 24 runs
Price source: Standard list-price estimate
Price list: boardgame-list-prices-2026-10-09
Average output: 9.6k tokens
Output range: 2.2k–23k tokens
Runs with output-token data: 24

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-6.1 Terra

Reasoning settings compared