CG-AL-X245

hard Codeunits & Business Logic content c6e2d9e1037a…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
GPT-6.1 Sol✓—100.0 / 1004.3k / 2.1k3
Gemini 3.8 Flash (OpenRouter)✓—100.0 / 1005.0k / 30k3
Claude Opus 5.5✓✓85.0 / 1008.3k / 2.7k3
GPT-6 Sol✓✓85.0 / 1004.8k / 1.7k3
Claude Opus 5✗✓81.3 / 1008.9k / 5.9k3
Claude Sonnet 5✗✓81.3 / 1009.9k / 9.0k3
GPT-5.6 Sol✗✓81.3 / 1005.0k / 2.0k3
GPT-5.6 Terra✗✓81.3 / 1005.1k / 2.3k3
GPT-5.6 Luna✗✓68.8 / 1006.9k / 3.3k3
Claude Haiku 4 5✗✗62.5 / 1009.5k / 1.1k3
GPT-6 Luna✗✓58.3 / 1005.6k / 2.3k3
GLM 5.3 (OpenRouter)✗✓56.3 / 1006.0k / 47k2
MiniMax M3✗✗52.1 / 1005.4k / 12k3