CG-AL-X239

hard Codeunits & Business Logic content afc256f506d3…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
Claude Opus 5✗✓81.3 / 1009.3k / 14k3
Claude Opus 5.5✗✓81.3 / 1009.1k / 3.4k3
Claude Sonnet 5✗✓81.3 / 1009.3k / 9.7k3
GPT-5.6 Luna✗✓81.3 / 1005.3k / 2.9k3
GPT-5.6 Terra✗✓81.3 / 1005.3k / 2.8k3
GPT-6.1 Sol✗✓81.3 / 1005.3k / 2.0k3
GPT-5.6 Sol✗✓75.0 / 1005.3k / 3.2k3
MiniMax M3✗✓75.0 / 1005.7k / 20k3
GPT-6 Luna✗✓68.8 / 1005.7k / 3.8k3
GPT-6 Sol✗✓68.8 / 1005.3k / 2.1k3
Gemini 3.8 Flash (OpenRouter)✗✓68.8 / 1006.3k / 40k3
Claude Haiku 4 5✗✗62.5 / 1007.9k / 1.1k3
GLM 5.3 (OpenRouter)✗✗15.6 / 1004.2k / 61k2