Model ledger

Claude Opus 5

Anthropic Claude Code · 11 recorded runs · 11 benchmarks. Open a run for its public headline result. Paid research access opens the full available ledger and configuration. Builder adds exact prompts, released projects, the RemakeBench Harness and production skills for the tests in its library.

Max reasoning

← All models Enter a blind battle

Benchmark

Fighter Game v1

1 run · Max reasoning

01Claude Opus 5 Max1h 23m 18.7s wall-clockWall-clock

Benchmark

JRPG Boss Battle

1 run · Max reasoning

01Claude Opus 5 Max1h 15m 49.9s wall-clockWall-clock

Benchmark

Jungle Temple

1 run · Max reasoning

01Claude Opus 5 Max52m 24.4s wall-clockWall-clock

Benchmark

MacBook-Class Cinematic

1 run · Max reasoning

01Claude Opus 5 Max1h 07m 02.5s wall-clockWall-clock

Benchmark

Neo-Gothic Storm City

1 run · Max reasoning

01Claude Opus 5 Max1h 52m 04.4s wall-clockWall-clock

Benchmark

Oasis Outpost

1 run · Max reasoning

01Claude Opus 5 Max49m 10.1s wall-clockWall-clock

Benchmark

Off-road Mud Game

1 run · Max reasoning

01Claude Opus 5 Max2h 48m 08.6s wall-clockWall-clock

Benchmark

Red Sands v1

1 run · Max reasoning

01Claude Opus 5 Max1h 30m 54.0s wall-clockWall-clock

Benchmark

Shrine Exploration

1 run · Max reasoning

01Claude Opus 5 Max7h 41m 36.5s wall-clockWall-clock

Benchmark

Space Flight Stage 2 — Landfall

1 run · Max reasoning

01Claude Opus 5 Max2h 31m 33.5s wall-clockWall-clock

Benchmark

Stained Glass Cathedral Shader

1 run · Max reasoning

01Claude Opus 5 Max1h 41m 22.5s wall-clockWall-clock