Get a program by ID Args: program_id: Program ID Returns: Program or None if not found

Add a program to the database Args: program: Program to add iteration: Current iteration (defaults to last_i

Run the evolution process with improved parallel processing Args: iterations: Maximum number of iterations (uses config

Load the database from disk Args: path: Path to load from

Get and format a fragment

Helper to create a test program

Helper to create a test program JSON file

Convert a value to float safely

Simulate what _submit_iteration does

Save the database to disk Args: path: Path to save to (uses config.db_path if None) iteration: Current itera

Calculate fitness score, excluding MAP-Elites feature dimensions This ensures that MAP-Elites features don't pollute the fitness calculation

Perform migration between islands This should be called periodically to share good solutions between islands

Close the tracer and flush remaining data

Calculate feature coordinates for the MAP-Elites grid Args: program: Program to calculate features for Returns:

Helper to create a test program

Mock evaluation that returns predictable metrics

Helper to create temporary evaluator file

Scale a feature value according to the configured scaling method Args: feature_name: Name of the feature dimension

Build a prompt for the LLM Args: current_program: Current program code parent_program: Parent program from w

Helper to create a test program with specific features

Extract evolution traces from an existing checkpoint directory Args: checkpoint_dir: Path to checkpoint directory output_pat

Retrieve all artifacts for a program Args: program_id: ID of the program Returns: Dictionary of art

Returns a standardized dictionary for failure cases.

Get a template by name

Log an evolution trace entry Args: iteration: Current iteration number parent_program: Parent program object

Sample a program and inspirations for the next evolution step Args: num_inspirations: Number of inspiration programs to

Submit an iteration to the process pool, optionally pinned to a specific island

Write buffered traces to file

Create a human-readable summary of the diff. For multi-line blocks, shows the full search and replace content (all lines). Args:

Store artifacts for a program Args: program_id: ID of the program artifacts: Dictionary of artifact name to

Process evaluation result to handle both dict and EvaluationResult returns Args: result: Raw result from evaluation func

Update statistics for a feature dimension Args: feature_name: Name of the feature dimension value: New value

Load configuration from a YAML file

Get and clear pending artifacts for a program Args: program_id: Program ID Returns: Artifacts dicti

Get the top N programs based on a metric Args: n: Number of programs to return metric: Metric to use for ran

Load configuration from a YAML file or use defaults

Helper to create evaluator with given settings (shorter timeout for faster tests)

Rebuild the models list after primary_model/secondary_model field changes

Run evolution with flexible inputs - the main library API Args: initial_program: Can be: - Path to a program file (str o

Fast approximation of code diversity using simple metrics Returns diversity score (higher = more diverse)

Get the best program based on a metric Args: metric: Metric to use for ranking (uses combined_score or average if None)

Calculate the average of numeric values in a metrics dictionary, safely ignoring non-numeric values like strings. Args: metrics:

Execute function with maximum GPU and Metal kernel protection

Calculate the bin index for a given complexity value using feature scaling. Args: complexity: The complexity value (code

Create comprehensive failure result with full error statistics

Helper to create a test program

Format a block of lines for diff summary: show all lines (truncated per line, optional cap).

Convert evaluator input to a file path

Evaluate a program and return scores Args: program_code: Code to evaluate program_id: Optional ID for loggin

Convert a value to float safely

Calculate the bin index for a given diversity value using feature scaling. Args: diversity: The average fast code divers

Determine the type/category of an inspiration program Args: program: Program dictionary Returns: St

Directly evaluate a program using the evaluation function with timeout Args: program_path: Path to the program file

Ensure clean GPU state before operations

Convert program input to a file path

Render artifacts for prompt inclusion Args: artifacts: Dictionary of artifact name to content Returns:

Resolve ${VAR} environment variable reference in a string value. In current implementation pattern must match the entire string (e.g., "${OPE

Safely calculate percentage improvement with bounds

Sample a completely random parent from all programs

Create comprehensive benchmark configurations

Safely format metrics dictionary for logging, handling both numeric and string values. Args: metrics: Dictionary of metric names to

Generate text using a system message and conversational context

Check if this result contains any artifacts

Log current status of all islands

Run a single benchmark configuration with proper warmup

Deserialize feature_stats from loaded JSON Args: stats_dict: Dictionary loaded from JSON Returns: P

Convert feature coordinates to a string key Args: coords: Feature coordinates Returns: String key

Format the inspirations section for the prompt Args: inspirations: List of inspiration programs language: Pr

Get diversity score for a program using cache and reference set Args: program: The program to calculate diversity for

Get config for library API tests with optillm server

Determine if program1 has better FITNESS than program2 Uses fitness calculation that excludes MAP-Elites feature dimensions

Safely decode an artifact value to string Args: value: Artifact value (string or bytes) Returns: St

Sample a model from the ensemble based on weights