Atlas

Benchmarks

← All benchmarks

Spanish

General QA · 2024-06-01

Elo rating on Scale SEAL's Spanish multilingual prompt set.

Top models (higher is better)

ModelScore
Gemini Exp-12061176
O11134
Gemini 2.0 Flash Experimental1119
o1 Preview1111
Gemini 2.0 Flash Thinking Experimental 01-211108
Gemini Experimental 11211105
GPT-4o1084
o3-mini1079
Gemini 1.5 Pro Preview 05141069
Gemini 1.5 Pro Experimental 08271067
GPT-4o (2024-08-06)1067
GPT-4 Turbo (1106 Preview)1034
Mistral Large 2 (Instruct 2407)1032
GPT-4 0125 Preview1020
Gemini 1.5 Pro Preview 04091005
Claude 3.5 Sonnet (June 2024)992
Aya Expanse 32B983
Gemini 1.5 Flash Preview 0514980
Gemma 2 27B951
Llama 3.2 90B Vision Instruct944
Opus 3919
Llama 3.1 405B Instruct915
Loading Atlas data…