Test fixture
Implementation-focused coding tasks with structured correctness checks.
The model receives the prompt (and optional system message). The run uses scorer javascript_function_tests with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Return JSON only with a string field named code. The code must be dependency-free JavaScript, define the requested function in the top level or module.exports, and include no markdown, imports, require, timers, network, filesystem, eval, or placeholders.
Implement function memoizeScript(script). The script is a comma-separated string where each token is 'call:N' meaning call an internal function with number argument N. The internal function returns N*N (a square). Use memoization so each unique argument is only computed once. Return an array: for each call operation, return the result. Also append the total number of actual computations performed (cache misses) as the last element.
Your JSON must look like {"code":"function memoizeScript(...) { ... }"}.{
"function_name": "memoizeScript",
"timeout_ms": 100,
"test_pass_threshold": 1,
"partial_credit_threshold": 0.5,
"tests": "[hidden executable tests]"
}temperature
0
max_tokens
1200
timeout (s)
120
type
scored
file
coding-medium-memoize.json