Benchmark menu

Model results

Claude Opus 4.5

See how Claude Opus 4.5 performed, which reasoning settings were tested, how often it produced a working player program, and what those runs cost.

Claude Opus 4.5

Provider: Anthropic

Rank
12
Status
Partial 3/3
Rank change
Baseline
Score
48.1
Last tested
2026-08-21
Settings tested
All settings tested

Why this result is partial: fixed opponent-and-seat coverage is incomplete. The rank uses the available reasoning-setting results.

Scores by reasoning setting

Available settings

Thinking field omittedLow effort + 8,000 thinking tokensMedium effort + 16,000 thinking tokensHigh effort + 32,000 thinking tokens

Checked on 2026-07-21 · Provider documentation

Scores from tested settings

Tested settingScoreResult statusDisplay group
Thinking field omitted49.0Tested and includedBaseline
Extended thinking (legacy shared budget)52.3Tested, not includedBalanced
Medium effort + 16,000 thinking tokens52.3Tested and includedBalanced
High effort + 32,000 thinking tokens43.0Tested and includedIntensive
Extended thinking (legacy shared budget)43.0Tested, not includedIntensive
Historical thinking-budget setting45.7Tested, not includedNot grouped

How settings are grouped

  • Baseline: Thinking field omitted — tested and included
  • Balanced: Medium effort + 16,000 thinking tokens — tested and included
  • Intensive: High effort + 32,000 thinking tokens — tested and included

Player program results

High effort + 32,000 thinking tokens: 100% / 0% / 0% · n 8Medium effort + 16,000 thinking tokens: 100% / 0% / 0% · n 8Thinking field omitted: 100% / 0% / 0% · n 8 All settings tested

Generation cost

Average estimated cost
$0.25
Median estimated cost
$0.21
Estimated range
$0.10–$0.51
Cost data
24 combinations / 40 recorded runs
Observed total
$11.20
Price source
Standard list-price estimate
Price list
boardgame-list-prices-2026-07-12 + boardgame-list-prices-2026-08-08
Average output
10.0k tokens
Output range
2.1k–28k tokens
Samples
24

Estimated cost gives equal weight to each model, game, and reasoning-setting combination.

Test setup

  • Combines this model's tested reasoning settings
  • Uses the standard GameBench player-program prompt