CG-AL-X264

hard Codeunits & Business Logic content 581134f19db6…

Description

Per-model results

Models that have attempted this task
ModelAttempt 1Attempt 2Avg scoreTokens / attemptRuns
GPT-6.1 Sol✓—100.0 / 1004.9k / 3.8k3
Gemini 3.8 Flash (OpenRouter)✓✓90.6 / 1005.8k / 32k3
GPT-6 Sol✓✓85.0 / 1005.0k / 2.2k3
Claude Opus 5.5✗✓81.3 / 1008.7k / 3.6k3
GPT-5.6 Sol✗✓81.3 / 1005.6k / 3.1k3
GPT-5.6 Terra✗✓81.3 / 1005.8k / 2.7k3
Claude Opus 5✗✓75.0 / 1009.7k / 16k3
Claude Sonnet 5✗✓68.8 / 10011k / 17k3
Claude Haiku 4 5✗✗62.5 / 1008.7k / 3.8k3
GPT-5.6 Luna✗✗62.5 / 1006.0k / 3.2k3
GPT-6 Luna✗✗62.5 / 1006.0k / 4.3k3
MiniMax M3✗✗31.3 / 1004.6k / 21k3
GLM 5.3 (OpenRouter)✗✗15.6 / 1004.6k / 61k2