DuelLabGameBench 2
Menu

Model results · Zhipu

GLM-5.3 Flash

Rank 23 of 66 in Overall Models. All three benchmark settings tested.

Score · relative 0–100
56.9
Rank
23
Status
Complete 3/3
Rank change
0
Players generated
2026-09-03
Match evidence updated
2026-10-26
Results published
Compare in charts View player programs Back to leaderboard

Scores by reasoning setting

  • Max (Intensive group, tested and included)59.1
  • High (Balanced group, tested and included)63.3
  • Low (Baseline group, tested and included)48.2

Every tested setting is included in the overall score.

Available settings and grouping

Available settings

LowHighMax

Checked on 2026-09-03 · Provider documentation

How settings are grouped

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

Scores by game

  • Go58.1
  • Chess52.6
  • Shogi71.7
  • Xiangqi72.5
  • Hex53.5
  • International Draughts47.5
  • Game of the Amazons72.4
  • Tumbleweed26.7

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 (63.3).

Across included settings, the highest game score is Xiangqi (72.5); the lowest is Tumbleweed (26.7). 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-flash
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
Not retained (0 of 28 indexed attempts)

Generation evidence

  • Low: 8 contributing player programs across 8 games
  • High: 8 contributing player programs across 8 games
  • Max: 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.

GameReasoning groupProgramsWorking-program strengthRaw ratingCombined record (W / D / L)
GoIntensive1 working of 1 attempt66.01627.6327 / 5 / 184
GoBalanced1 working of 1 attempt71.11684.0346 / 20 / 150
GoBaseline1 working of 1 attempt37.31346.574 / 217 / 225
ChessIntensive1 working of 1 attempt66.41607.2188 / 246 / 46
ChessBalanced1 working of 1 attempt46.41443.940 / 349 / 91
ChessBaseline1 working of 1 attempt44.91432.526 / 363 / 91
ShogiIntensive1 working of 1 attempt78.41739.5369 / 8 / 117
ShogiBalanced1 working of 1 attempt80.51774.2372 / 25 / 97
ShogiBaseline1 working of 1 attempt56.41485.3202 / 114 / 178
XiangqiIntensive1 working of 1 attempt78.81739.3317 / 138 / 43
XiangqiBalanced1 working of 1 attempt74.01684.0281 / 160 / 57
XiangqiBaseline1 working of 1 attempt64.61593.5212 / 202 / 84
HexIntensive1 working of 1 attempt74.31789.2385 / 0 / 145
HexBalanced1 working of 1 attempt48.61419.6249 / 0 / 281
HexBaseline1 working of 1 attempt37.61280.1192 / 0 / 338
International DraughtsIntensive1 working of 1 attempt17.2914.166 / 14 / 422
International DraughtsBalanced1 working of 1 attempt81.51973.9383 / 32 / 87
International DraughtsBaseline1 working of 1 attempt43.91325.5188 / 18 / 295
Game of the AmazonsIntensive1 working of 1 attempt69.71713.3362 / 0 / 168
Game of the AmazonsBalanced1 working of 1 attempt75.61809.9396 / 0 / 134
Game of the AmazonsBaseline1 working of 1 attempt71.81746.0374 / 0 / 156
TumbleweedIntensive1 working of 1 attempt21.91077.186 / 44 / 408
TumbleweedBalanced1 working of 1 attempt29.11160.184 / 120 / 334
TumbleweedBaseline1 working of 1 attempt29.11160.197 / 94 / 347

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: 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

Max: 88% / 12% / 0% · n 8High: 100% / 0% / 0% · n 8Low: 63% / 37% / 0% · n 8 All three benchmark settings tested

Generation cost

$0.0081average estimated cost per game and reasoning setting

Median
$0.0050
Range
$0.0011–$0.03
Provider-reported total
$0.20
Cost data
24 combinations · 24 runs
Price source
Provider-reported
Price list
Not recorded for this older data
Average output
27k tokens
Output range
2.3k–66k 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