CG-AL-X290

hard Codeunits & Business Logic content c2cf49341d8a…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
Claude Opus 5.5✓—100.0 / 1009.1k / 7.2k3
GPT-6.1 Sol✓✓85.0 / 1005.7k / 3.0k3
Claude Opus 5✗✓81.3 / 10010k / 8.4k3
GPT-5.6 Luna✗✓81.3 / 1005.9k / 2.5k3
GPT-5.6 Terra✗✓81.3 / 1006.5k / 3.8k3
GPT-6 Luna✗✓81.3 / 1005.9k / 3.0k3
Gemini 3.8 Flash (OpenRouter)✓✗81.3 / 1006.1k / 35k3
GPT-6 Sol✗✓75.0 / 1006.2k / 2.5k3
MiniMax M3✗✓68.8 / 1007.2k / 17k3
Claude Sonnet 5✗✓64.6 / 10011k / 13k3
GPT-5.6 Sol✗✗52.1 / 1006.2k / 4.2k3
GLM 5.3 (OpenRouter)✗✓40.6 / 1004.5k / 57k2
Claude Haiku 4 5✗✗20.8 / 1006.9k / 1.7k3