List your company

Climate Discovery on ClimaBench 2000–2024

7.38AE (Absolute Error)DeepSeek-R1
7.28927.90218.5159.1279Apr 18, 2026

Evaluation results

MethodLinks
DeepSeek-R1Backbone=DeepSeek-R1-671B2026.047.384.888.754.626.41
Qwen3-235BBackbone=Qwen3-235B2026.047.424.928.724.686.45
GPT-4oBackbone=GPT-4o2026.047.553.928.425.256.29
ResearchAgentBackbone=GPT-4o2026.048.056.858.857.427.79
DS-AgentBackbone=GPT-4o2026.048.247.128.757.517.91
ResearchAgentBackbone=DeepSeek-R1-671B2026.048.247.218.857.127.86
ResearchAgentBackbone=Qwen3-235B2026.048.287.258.887.157.89
DS-AgentBackbone=DeepSeek-R1-671B2026.048.357.028.827.357.89
DS-AgentBackbone=Qwen3-235B2026.048.387.058.857.387.92
Agent LaboratoryBackbone=GPT-4o2026.048.626.428.685.687.35
DeepAnalyzeBackbone=GPT-4o2026.048.626.288.756.227.34
DeepAnalyzeBackbone=DeepSeek-R1-671B2026.048.626.288.756.227.34
DeepAnalyzeBackbone=Qwen3-235B2026.048.626.288.756.227.34
Agent LaboratoryBackbone=DeepSeek-R1-671B2026.048.756.128.86.127.45
Agent LaboratoryBackbone=Qwen3-235B2026.048.786.158.826.157.48
ClimAgentBackbone=GPT-4o2026.049.227.459.128.588.92
ClimAgentBackbone=DeepSeek-R1-671B2026.049.628.359.188.658.95
ClimAgentBackbone=Qwen3-235B2026.049.658.389.228.688.98