CG-AL-X254

hard Codeunits & Business Logic content 4ddef16d31e9…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
Claude Opus 5.5✓—100.0 / 10010k / 6.3k3
Gemini 3.8 Flash (OpenRouter)✓—100.0 / 1006.5k / 29k3
GPT-6.1 Sol✓✓90.6 / 1006.1k / 2.2k3
Claude Opus 5✗✓81.3 / 10011k / 6.5k3
Claude Sonnet 5✗✓81.3 / 10011k / 10k3
GLM 5.3 (OpenRouter)✗✓81.3 / 1006.6k / 45k2
GPT-5.6 Sol✗✗62.5 / 1007.4k / 4.1k3
GPT-5.6 Terra✗✗62.5 / 1007.3k / 3.2k3
GPT-6 Sol✗✗62.5 / 1006.8k / 2.8k3
MiniMax M3✗✓54.2 / 1005.2k / 10k3
Claude Haiku 4 5✗✗52.1 / 10012k / 1.6k3
GPT-6 Luna✗✗41.7 / 1006.3k / 3.5k3
GPT-5.6 Luna✗✗31.3 / 1006.8k / 2.7k3