Score · relative 0–100
43.1
Rank
61
Status
Complete 3/3
Rank change
New
Players generated
2026-11-16
Match evidence updated
2026-11-19

Reasoning settings compared

  1. MaxIntensive

    40.1 · rank 69 on the Intensive leaderboard

    Cost per program
    $0.0409 · provider-reported
    Working programs
    7 of 8
    API time
    1.8 min (observed mean, descriptive)

    6.1 points below Medium, at 3.1× the cost

    • Higher score for $0.0409 or less: Pareto 26.10 (no Overall rank) Reasoning controls unavailable · Baseline: 89.9 for $0.0296 (provider-reported)
    • Cheapest scoring at least as high: Muse Ember Minimal · Baseline: 46.9 for $0.00281 (provider-reported) (one of 39 such settings)
  2. MediumBalanced

    46.2 · rank 64 on the Balanced leaderboard

    Cost per program
    $0.0131 · provider-reported
    Working programs
    8 of 8
    API time
    28 s (observed mean, descriptive)

    3.3 points above Minimal, at 1.5× the cost

    • Higher score for $0.0131 or less: Particle-1 High · Balanced: 62.8 for $0.0130 (provider-reported)
    • Cheapest scoring at least as high: Muse Ember Minimal · Baseline: 46.9 for $0.00281 (provider-reported) (one of 14 such settings)
  3. MinimalBaseline

    42.9 · rank 51 on the Baseline leaderboard

    Cost per program
    $0.00897 · provider-reported
    Working programs
    8 of 8
    API time
    11 s (observed mean, descriptive)
    • Higher score for $0.00897 or less: GLM-5.3 Flash High · Balanced: 60.5 for $0.00579 (provider-reported)
    • Cheapest scoring at least as high: Muse Ember Minimal · Baseline: 46.9 for $0.00281 (provider-reported) (one of 10 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

MinimalLowMediumHighMax

Checked on 2026-11-16 · Provider documentation

How settings are grouped

  • Baseline: Minimal — tested and included
  • Balanced: Medium — tested and included
  • Intensive: Max — tested and included

Scores by game

  • Go55.1
  • Chess1 of 3 failed28.5
  • Shogi35.0
  • Xiangqi40.1
  • Hex43.9
  • International Draughts33.2
  • Game of the Amazons55.2
  • Tumbleweed53.7

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: Medium (46.2).

Across included settings, the highest game score is Game of the Amazons (55.2); the lowest is Chess (28.5). 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
thinkingmachines/inkling-2-small
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
943718 (28 of 28 indexed attempts)

Generation evidence

  • Minimal: 8 contributing player programs across 8 games
  • Medium: 8 contributing player programs across 8 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.

GameReasoning groupProgramsWorking-program strengthRaw ratingCombined record (W / D / L)
GoIntensive1 working of 1 attempt55.11496.8174 / 317 / 153
GoBalanced1 working of 1 attempt66.91616.3320 / 180 / 148
GoBaseline1 working of 1 attempt43.31383.1116 / 287 / 241
ChessBalanced1 working of 1 attempt42.11374.78 / 447 / 155
ChessBaseline1 working of 1 attempt43.31384.519 / 442 / 153
ShogiIntensive1 working of 1 attempt31.71209.998 / 144 / 392
ShogiBalanced1 working of 1 attempt45.01329.2175 / 142 / 317
ShogiBaseline1 working of 1 attempt28.21175.359 / 182 / 393
XiangqiIntensive1 working of 1 attempt41.41377.3130 / 215 / 283
XiangqiBalanced1 working of 1 attempt40.21367.667 / 327 / 234
XiangqiBaseline1 working of 1 attempt38.71355.338 / 366 / 224
HexIntensive1 working of 1 attempt64.91552.2395 / 0 / 273
HexBalanced1 working of 1 attempt20.1906.8119 / 0 / 543
HexBaseline1 working of 1 attempt46.71302.7283 / 0 / 379
International DraughtsIntensive1 working of 1 attempt10.3725.453 / 2 / 579
International DraughtsBalanced1 working of 1 attempt43.61294.5241 / 6 / 385
International DraughtsBaseline1 working of 1 attempt45.61326.0245 / 23 / 366
Game of the AmazonsIntensive1 working of 1 attempt47.71333.1275 / 0 / 387
Game of the AmazonsBalanced1 working of 1 attempt72.91693.4447 / 0 / 215
Game of the AmazonsBaseline1 working of 1 attempt44.91297.1257 / 0 / 405
TumbleweedIntensive1 working of 1 attempt69.71728.6432 / 0 / 234
TumbleweedBalanced1 working of 1 attempt38.71159.9233 / 0 / 429
TumbleweedBaseline1 working of 1 attempt52.81385.3323 / 0 / 339

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: 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 8Medium: 100% / 0% / 0% · n 8Minimal: 75% / 25% / 0% · n 8 All three benchmark settings tested

Generation cost, all settings combined

$0.02average estimated cost per game and reasoning setting

Median
$0.01
Range
$0.0057–$0.06
Provider-reported total
$0.50
Cost data
24 combinations · 24 runs
Price source
Provider-reported
Price list
boardgame-list-prices-2026-11-12
Average output
18k tokens
Output range
3.1k–55k 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