CG-AL-X274

hard Codeunits & Business Logic content 1a98fe2ba332…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
GPT-5.6 Sol✗✓81.3 / 1006.5k / 3.8k3
GPT-5.6 Terra✗✓81.3 / 1006.8k / 3.5k3
GPT-6.1 Sol✗✓81.3 / 1006.0k / 2.8k3
Gemini 3.8 Flash (OpenRouter)✗✓81.3 / 1007.1k / 25k3
Claude Opus 5✗✓75.0 / 10012k / 11k3
Claude Sonnet 5✗✓75.0 / 10012k / 12k3
GPT-6 Sol✗✓75.0 / 1006.4k / 2.8k3
Claude Opus 5.5✗✓68.8 / 10011k / 5.7k3
Claude Haiku 4 5✗✗62.5 / 1009.7k / 3.8k3
GPT-5.6 Luna✗✓58.3 / 1008.4k / 3.2k3
GPT-6 Luna✗✓47.9 / 1007.7k / 4.2k3
MiniMax M3✗✗20.8 / 1003.6k / 13k3
GLM 5.3 (OpenRouter)✗✗15.6 / 1005.3k / 59k2