Scores by reasoning setting

Enabled (Balanced group, tested and included)40.6

Every tested setting is included in the overall score.

Available settings and grouping

Available settings

Reasoning enabled

Checked on 2026-07-21 · Provider documentation

How settings are grouped

Baseline: Not available for this model
Balanced: Reasoning enabled — tested and included
Intensive: Not available for this model

Scores by game

Go38.5
Chess42.6
Shogiexcluded: provider, API or test-machine fault, not counted—
Xiangqi38.4
Hex53.0
International Draughts1 of 1 failed0.0
Game of the Amazons43.5
Tumbleweed68.3

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: Reasoning enabled (40.6).

Across included settings, the highest game score is Tumbleweed (68.3); the lowest is International Draughts (0.0). 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: moonshotai/kimi-k2.7-code
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 18 indexed attempts)

Generation evidence

Reasoning enabled: 6 contributing player programs across 6 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	Balanced	1 working of 1 attempt	38.5	1358.9	185 / 7 / 323
Chess	Balanced	1 working of 1 attempt	42.6	1414.5	15 / 363 / 102
Xiangqi	Balanced	1 working of 1 attempt	38.4	1384.9	38 / 291 / 169
Hex	Balanced	1 working of 1 attempt	53.0	1474.5	272 / 0 / 258
Game of the Amazons	Balanced	1 working of 1 attempt	43.5	1366.9	217 / 0 / 313
Tumbleweed	Balanced	1 working of 1 attempt	68.3	1655.2	333 / 33 / 172

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; 1 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

Reasoning enabled: 0% / 86% / 14% · n 7 No Overall rankIntensive not supportedBaseline not supportedMany programs needed repair

Generation cost

$0.16average estimated cost per game and reasoning setting

Median: $0.14
Range: $0.11–$0.23
Known cost subtotal (incomplete or unverified charges): $1.28
Cost data: 8 combinations · 8 runs
Price source: Provider-reported
Price list: Not recorded for this older data
Average output: 39k tokens
Output range: 23k–56k tokens
Runs with output-token data: 8

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

Kimi K2.7 Code