CG-AL-X298

hard Codeunits & Business Logic content 38ce3a3472b7…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
GPT-6.1 Sol✓—100.0 / 1004.4k / 1.7k3
GPT-5.6 Sol✓✓90.6 / 1004.5k / 3.5k3
GPT-5.6 Terra✓✓90.6 / 1004.5k / 3.0k3
GPT-6 Sol✓✓90.6 / 1004.5k / 2.6k3
Gemini 3.8 Flash (OpenRouter)✓✓85.0 / 1006.4k / 23k3
Claude Opus 5.5✓✗81.3 / 1007.9k / 4.2k3
Claude Sonnet 5✗✗62.5 / 1007.9k / 23k3
GPT-5.6 Luna✗✗62.5 / 1007.0k / 3.2k3
GPT-6 Luna✗✗62.5 / 1005.6k / 4.1k3
Claude Opus 5✓✓52.5 / 1007.7k / 12k3
MiniMax M3✗✗41.7 / 1004.9k / 19k3
Claude Haiku 4 5✗✗31.3 / 1009.7k / 2.9k3
GLM 5.3 (OpenRouter)✗✗0.0 / 1002.7k / 64k2