CG-AL-X289

hard Codeunits & Business Logic content b049efd035b1…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
Claude Opus 5.5✓—100.0 / 1008.8k / 9.3k3
GPT-6.1 Sol✓—100.0 / 1005.0k / 4.9k3
Gemini 3.8 Flash (OpenRouter)✓✓90.6 / 1006.0k / 38k3
Claude Opus 5✗✓81.3 / 10010k / 19k3
Claude Sonnet 5✗✓81.3 / 10010k / 14k3
GPT-5.6 Sol✗✓64.6 / 1005.8k / 4.0k3
GPT-5.6 Terra✗✓64.6 / 1005.8k / 3.5k3
Claude Haiku 4 5✗✗62.5 / 1009.7k / 2.9k3
GPT-5.6 Luna✗✓58.3 / 1007.4k / 4.0k3
GPT-6 Luna✗✓47.9 / 1005.6k / 5.3k3
GPT-6 Sol✗✗20.8 / 1005.2k / 3.1k3
MiniMax M3✗✗0.0 / 100491 / 16k3
GLM 5.3 (OpenRouter)✗✗0.0 / 1003.0k / 64k2