List your company

Multiple-choice Question Answering on WeatherQA (test)

44.8Accuracy (500 hPa)Weather-R1-7B
34.08836.86939.6542.431Jan 20, 2026

Evaluation results

MethodLinks
Weather-R1-7BEvaluation Protocol=Fine-tuned, Model Scale=7B, Optimization=LoCo-RFT2026.0144.847.546.641.964.467.757.452.9
SFT-7BEvaluation Protocol=Fine-tuned, Model Scale=7B2026.0143.842.949.940.66258.450.849.8
RFT-7BEvaluation Protocol=Fine-tuned, Model Scale=7B2026.0143.148.447.941.564.662.951.451.4
Qwen2.5-VL-3BEvaluation Protocol=Zero-shot, Model Scale=3B2026.013845.53623.552.5604442.8
LLaVA-v1.6-13BEvaluation Protocol=Zero-shot, Model Scale=13B2026.0135.53528226023.52532.7
Qwen2.5-VL-7BEvaluation Protocol=Zero-shot, Model Scale=7B2026.0135.539.530.5355756.547.543.1
LLaVA-v1.6-7BEvaluation Protocol=Zero-shot, Model Scale=7B2026.0134.526.5233727.5252027.6