List your company

Geological Reasoning on Geo-Eval

7.35Concept ScoreGPT-5.4 (Frontier)
4.1785.00155.8256.6485May 24, 2026

Evaluation results

MethodLinks
GPT-5.4 (Frontier)Size=-2026.057.357.177.15—
DeepSeek-V3.2-InstructSize=-2026.056.86.756.676.74—
Qwen3-32B-geoSize=32B2026.056.786.796.96.821.82
Gemma-3-27B-geoSize=27B2026.056.76.66.476.591.43
GPT-4oSize=-2026.056.15.95.85.93—
Qwen3-8B-geoSize=8B2026.056.16.276.446.271.64
Gemma-3-27B-ITSize=27B2026.055.35.15.085.16—
Qwen3-32BSize=32B2026.055.24.94.95—
Qwen3-8BSize=8B2026.054.84.684.414.63—
GLM-4-9BSize=9B2026.054.74.54.514.57—
Llama-3.1-70B-InstructSize=70B2026.054.34.13.964.12—