Skip to content

DSPy.LM improvements: we've added a typed BaseLM that will become the default interface. Read the plan and share feedback

DSPy

Evaluation

Initializing search

stanfordnlp/dspy

Community
FAQ

Overview
Getting Started
Diving Deeper
Tutorials
API Reference

DSPy

stanfordnlp/dspy

Overview
Getting Started
Getting Started
Diving Deeper
Diving Deeper
Tutorials
Tutorials
- Build AI Programs with DSPy
  Build AI Programs with DSPy
- Optimize AI Programs with DSPy
  Optimize AI Programs with DSPy
- Reflective Prompt Evolution with dspy.GEPA
  Reflective Prompt Evolution with dspy.GEPA
- Experimental RL Optimization for DSPy
  Experimental RL Optimization for DSPy
  - RL for Privacy-Conscious Delegation
  - RL for Multi-Hop Research
- Tools, Development, and Deployment
  Tools, Development, and Deployment
- Real-World Examples
  Real-World Examples
Community
Community
FAQ
FAQ
- FAQ
- Cheatsheet
API Reference
API Reference
- Adapters
  Adapters
- Evaluation
  Evaluation
- Experimental
  Experimental
  - Citations
  - Document
- Models
  Models
  - BaseLM
  - Embedder
  - LM
- Modules
  Modules
  - BestOfN
  - ChainOfThought
  - CodeAct
  - Module
  - MultiChainComparison
  - Parallel
  - Predict
  - ProgramOfThought
  - ReAct
  - Refine
  - RLM
- Optimizers
  Optimizers
  - GEPA
    GEPA
    
    1. GEPA Overview
    
    2. GEPA Advanced
  - BetterTogether
  - BootstrapFewShot
  - BootstrapFewShotWithRandomSearch
  - BootstrapFinetune
  - BootstrapRS
  - COPRO
  - Ensemble
  - InferRules
  - KNN
  - KNNFewShot
  - LabeledFewShot
  - MIPROv2
  - SIMBA
- Primitives
  Primitives
  - Audio
  - Code
  - Example
  - History
  - Image
  - Prediction
  - Tool
  - ToolCalls
- Signatures
  Signatures
- Tools
  Tools
- Utils
  Utils

Overview
API Reference
Evaluation

Evaluation¶

API reference for DSPy evaluation. Select a page below.

answer_exact_match
answer_passage_match
CompleteAndGrounded
Evaluate
EvaluationResult
SemanticF1

CompleteAndGrounded

© 2026 DSPy

Made with Material for MkDocs