Score · relative 0–100: 57.1
Rank: 23
Status: Complete 3/3
Rank change: 0
Players generated: 2026-11-03
Match evidence updated: 2026-11-10

Compare in charts View player programs Back to leaderboard

MaxIntensive
69.8 · rank 21 on the Intensive leaderboard
Cost per program
$0.105 · provider-reported
Working programs
7 of 8
API time
9.1 min (observed mean, descriptive)
10.9 points above High, at 1.4× the cost
- Higher score for $0.105 or less: Pareto 26.10 (no Overall rank) Reasoning controls unavailable · Baseline: 89.7 for $0.0296 (provider-reported)
- Cheapest scoring at least as high: Particle-1 Max · Intensive: 71.7 for $0.0256 (provider-reported) (one of 14 such settings)
HighBalanced
58.9 · rank 28 on the Balanced leaderboard
Cost per program
$0.0767 · provider-reported
Working programs
7 of 8
API time
5.1 min (observed mean, descriptive)
16.3 points above Low, at 6.0× the cost
- Higher score for $0.0767 or less: Pareto 26.10 (no Overall rank) Reasoning controls unavailable · Baseline: 89.7 for $0.0296 (provider-reported)
- Cheapest scoring at least as high: GLM-5.3 Flash High · Balanced: 60.8 for $0.00579 (provider-reported) (one of 19 such settings)
LowBaseline
42.6 · rank 38 on the Baseline leaderboard
Cost per program
$0.0128 · provider-reported
Working programs
7 of 8
API time
37 s (observed mean, descriptive)
- Higher score for $0.0128 or less: GLM-5.3 Flash High · Balanced: 60.8 for $0.00579 (provider-reported)
- Cheapest scoring at least as high: GPT-6 Luna None · Baseline: 45.3 for $0.00318 (list-price estimate) (one of 12 such settings)

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

LowHighMax

Checked on 2026-11-03 · Provider documentation

How settings are grouped

Baseline: Low — tested and included
Balanced: High — tested and included
Intensive: Max — tested and included

Scores by game

Go1 of 3 failed36.1
Chess65.3
Shogi67.4
Xiangqi2 of 3 failed18.3
Hex74.6
International Draughts66.4
Game of the Amazons70.1
Tumbleweed58.6

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: Max (69.8).

Across included settings, the highest game score is Hex (74.6); the lowest is Xiangqi (18.3). 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: z-ai/glm-5.3-flashx-legal
Provider route (catalog): openrouter
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: 131072 (33 of 33 indexed attempts)

Generation evidence

Low: 7 contributing player programs across 7 games
High: 7 contributing player programs across 7 games
Max: 7 contributing player programs across 7 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	66.8	1649.2	357 / 31 / 176
Go	Balanced	1 working of 1 attempt	41.6	1399.2	138 / 184 / 242
Chess	Intensive	1 working of 1 attempt	82.1	1782.8	390 / 81 / 61
Chess	Balanced	1 working of 1 attempt	59.6	1547.4	156 / 304 / 72
Chess	Baseline	1 working of 1 attempt	54.1	1502.4	85 / 387 / 60
Shogi	Intensive	1 working of 1 attempt	83.7	1852.3	446 / 5 / 97
Shogi	Balanced	1 working of 1 attempt	81.3	1803.5	432 / 3 / 113
Shogi	Baseline	1 working of 1 attempt	37.2	1315.7	107 / 158 / 283
Xiangqi	Baseline	1 working of 1 attempt	54.9	1509.0	206 / 165 / 173
Hex	Intensive	1 working of 1 attempt	93.3	2433.0	540 / 0 / 42
Hex	Balanced	1 working of 1 attempt	74.0	1798.0	420 / 0 / 162
Hex	Baseline	1 working of 1 attempt	56.4	1505.2	317 / 0 / 265
International Draughts	Intensive	1 working of 1 attempt	85.9	2108.1	469 / 6 / 79
International Draughts	Balanced	1 working of 1 attempt	50.1	1435.5	252 / 16 / 286
International Draughts	Baseline	1 working of 1 attempt	63.2	1659.1	334 / 6 / 214
Game of the Amazons	Intensive	1 working of 1 attempt	76.9	1835.0	439 / 0 / 139
Game of the Amazons	Balanced	1 working of 1 attempt	74.9	1798.8	427 / 0 / 151
Game of the Amazons	Baseline	1 working of 1 attempt	58.5	1544.6	324 / 0 / 254
Tumbleweed	Intensive	1 working of 1 attempt	69.9	1791.2	396 / 0 / 186
Tumbleweed	Balanced	1 working of 1 attempt	89.2	2263.0	515 / 0 / 67
Tumbleweed	Baseline	1 working of 1 attempt	16.7	888.5	90 / 0 / 492

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

3 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

Max: 75% / 13% / 12% · n 8High: 75% / 13% / 12% · n 8Low: 38% / 50% / 12% · n 8 All three benchmark settings testedMany programs needed repair

Generation cost, all settings combined

$0.06average estimated cost per game and reasoning setting

Median: $0.07
Range: $0.0066–$0.24
Provider-reported total: $1.55
Cost data: 24 combinations · 24 runs
Price source: Provider-reported
Price list: boardgame-list-prices-2026-10-09
Average output: 43k tokens
Output range: 3.2k–103k 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

GLM-5.3 FlashX Legal

Reasoning settings compared