Benchmark · agentic

GDPval-AA (Elo)

2 resultados 2 modelos

Artificial Analysis's pairwise-graded Elo ranking on real-world economically valuable knowledge work; four-digit scores (~640-1950 Elo), so never file a percentage here. Not Chatbot Arena Elo (lmsys-arena) and not OpenAI's percentage GDPval.

1000 1150 1300 1450 1600 2026-07-27 Gemini 3.6 Flash · 1421 · 2026-07-27 Gemini 3.5 Flash-Lite · 1140 · 2026-07-27
Gemini 3.6 Flash Gemini 3.5 Flash-Lite
Cronología
Fecha Modelo Puntuación Fuente
2026-07-27 Gemini 3.6 Flash 1421.0Elo Google lanza Gemini 3.6 Flash, 3.5 Flash-Lite y 3.5 Flash Cyber
2026-07-27 Gemini 3.5 Flash-Lite 1140.0Elo Google lanza Gemini 3.6 Flash, 3.5 Flash-Lite y 3.5 Flash Cyber