promptfoo--promptfoo/examples/eval-bert-score/promptfooconfig-advanced.yaml

# yaml-language-server: $schema=https://promptfoo.dev/config-schema.json
description: Advanced BERTScore with multiple references

prompts:
  - 'Explain: {{topic}}'

providers:
  - openai:gpt-4.1-nano

tests:
  - vars:
      topic: 'gradient descent'
      reference: 'An optimization algorithm that adjusts parameters to minimize error'
    assert:
      - type: python
        value: |
          # Compare against multiple valid answers
          from bert_score import score

          references = [
              context['vars']['reference'],
              "A method for finding the minimum of a function by moving in the direction of steepest descent",
              "Like rolling a ball down a hill to find the lowest point"
          ]

          # Get best score across all references
          scores = []
          for ref in references:
              _, _, F1 = score([output], [ref], lang='en', verbose=False)
              scores.append(F1.item())

          return max(scores)