CG-AL-X296

hard Codeunits & Business Logic content 4303d6be1a84…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
GPT-6.1 Sol✓—100.0 / 1006.6k / 4.6k3
GPT-5.6 Sol✓✓90.6 / 1008.2k / 6.6k3
Claude Opus 5✗✓81.3 / 10013k / 14k3
Claude Opus 5.5✗✓81.3 / 10012k / 6.9k3
GPT-6 Luna✗✓81.3 / 1008.5k / 5.0k3
GPT-6 Sol✗✓81.3 / 1007.0k / 3.8k3
Gemini 3.8 Flash (OpenRouter)✗✓81.3 / 1008.0k / 28k3
Claude Haiku 4 5✗✗62.5 / 10010k / 3.0k3
Claude Sonnet 5✗✗62.5 / 10015k / 13k3
GPT-5.6 Luna✗✗62.5 / 1008.4k / 3.8k3
MiniMax M3✗✗62.5 / 1008.7k / 23k3
GPT-5.6 Terra✗✗41.7 / 1007.7k / 4.8k3
GLM 5.3 (OpenRouter)✗✗31.3 / 1008.1k / 53k2