benchgap
Calibration

Gert Labs → MCP Atlas

MCP Atlas is estimated from Gert Labs with a offset logistic curve fitted on 13 models measured on both: y = 0.4354 + (0.7175 − 0.4354) / (1 + exp(−200.00·(x − 0.5694))), R² = 0.55, cross-validated error 14.8 pp. It is used for 7 estimates.

Estimated modelGert LabsMCP AtlasSource
Gemini 3.1 Flash-Lite38.5%43.5%estimated ± 14.8 pp, low confidence
GPT-4.125.7%43.5%estimated ± 14.8 pp, low confidence
Grok 442.3%43.5%estimated ± 14.8 pp, low confidence
Grok 4.1 Fast47.3%43.5%estimated ± 14.8 pp, low confidence
Grok 4.2038.4%43.5%estimated ± 14.8 pp, low confidence
Grok Build 0.149.2%43.5%estimated ± 14.8 pp, low confidence
Qwen3 Max43.7%43.5%estimated ± 14.8 pp, low confidence