List your company
ResearchFormerly WizWand

Well-being support evaluation on 10 situational well-being prompts (test)

9.5ScoreClaude 3.7 Sonnet
3.3644.9576.558.143Nov 18, 2025
Updated 4mo ago

Evaluation results

MethodLinks
Claude 3.7 SonnetJudge=GPT-52025.119.5
GPT-4oJudge=GPT-4o2025.118.9
GPT-4oJudge=GPT-52025.118.8
Claude 3.7 SonnetJudge=GPT-4o2025.118.7
Gemma-3Judge=GPT-4o2025.118.7
LLaMA-3Judge=GPT-52025.118.6
LLaMA-3Judge=GPT-4o2025.118.6
Al LunaJudge=GPT-52025.118.5
Gemma-3Judge=GPT-52025.118.5
Al LunaJudge=GPT-4o2025.118.5
LlamaSupportJudge=GPT-52025.118.4
LlamaSupportJudge=GPT-4o2025.118.4
MistralJudge=GPT-4o2025.118.3
MistralJudge=GPT-52025.117.9
Phi-4Judge=GPT-52025.117.3
Phi-4Judge=GPT-4o2025.117.3
MentalLLaMA-2Judge=GPT-4o2025.116.5
MentalLLaMA-2Judge=GPT-52025.115.8
Tell MeJudge=Human, Pipeline=RAG Pipeline2025.113.8
Tell MeJudge=Human, Pipeline=Non-RAG (Baseline)2025.113.6