SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.15877 · ICLR · 2025

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

Qian Liu, Chen Gong, Jean Kaddour, Wenhao Yu, Han Hu, Alex Gu, Daniel Fried, Jenny Chim, Minh Chien Vu, Zhoujun Cheng, Zijian Wang, Leandro Von Werra, and 22 more

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2406.15877.json · badge: /api/ran/2406.15877.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
bigcode-project/bigcodebench-annotation canonical (classifier label)
classifier basis: census
3 ran · 1 no recorded run
bigcode-project/bigcodebench canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
extract_apis Ran
ran; our contract draft was wrong, not the code
bigcode-project/bigcodebench-annotation/script/parse.py
code served (permissive licence) · get_code("ff8aaee37e35336e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import ast

# The function takes a 'code' parameter which should be a string containing Python code
# The function parses it and extracts API calls using ast.NodeVisitor

# Create a simple test code string with various imports and API usage
test_code = """
import numpy as np
from torch import nn
import torch.optim as optim

class MyModel:
    def __init__(self):
        self.layer = nn.Linear(10, 5)
    
    def forward(self):
        x = np.array([1, 2, 3])
        optimizer = optim.Adam([])
        return x

result = np.mean([1, 2, 3])
"""

ARGS = {
    "code": test_code
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
filter_unused_imports Ran
ran; our contract draft was wrong, not the code
bigcode-project/bigcodebench-annotation/script/parse.py
code served (permissive licence) · get_code("8f5519c167210a90")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

# Construct valid arguments for filter_unused_imports
script = """import os
import sys
from numpy import array
import torch
from collections import defaultdict

def my_function():
    x = os.path.exists('file.txt')
    y = array([1, 2, 3])
    return x, y
"""

libs = ["numpy", "torch", "os"]

ARGS = {
    "script": script,
    "libs": libs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_function_dependency Ran
ran; our contract draft was wrong, not the code
bigcode-project/bigcodebench/bigcodebench/sanitize.py
code served (permissive licence) · get_code("5f0eb74fd5059693")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned set. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from typing import Dict, Generator, List, Optional, Set, Tuple

def get_function_dependency(entrypoint: str, call_graph: Dict[str, str]) -> Set[str]:
    queue = [entrypoint]
    visited = {entrypoint}
    while queue:
        current = queue.pop(0)
        if current not in call_graph:
            continue
        for neighbour in call_graph[current]:
            if not neighbour in visited:
                visited.add(neighbour)
                queue.append(neighbour)
    return visited

ARGS = {
    "entrypoint": "func_a",
    "call_graph": {
        "func_a": ["func_b", "func_c"],
        "func_b": ["func_d"],
        "func_c": ["func_d", "func_e"],
        "func_d": [],
        "func_e": ["func_f"],
        "func_f": []
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
remove_trailing_comments Ran
ran; our contract draft was wrong, not the code
bigcode-project/bigcodebench-annotation/script/parse.py
code served (permissive licence) · get_code("881d446392c3564a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

ARGS = {
    'text': """def hello_world():
    print("Hello, World!")
# This is a trailing comment
# Another trailing comment
"""
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
syntax_check Ran
ran, violated the contract we drafted
bigcode-project/bigcodebench/bigcodebench/syncheck.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from GATECH-EIC/R2-dLLM, is pointer only (licence: NONE) · get_code("d7d30a99e0029e36")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from GATECH-EIC/R2-dLLM, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_pypi_stats Not yet run
unverified (harvested, no recorded run)
bigcode-project/bigcodebench-annotation/analysis/pypi_download_stats.py
code served (permissive licence) · get_code("c1ac61d6490eade0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
syntax_check 5 other papers
5 later
2412.03578 (2024-12)
PerfCodeGen: Improving Performance of LLM Generated Code with Execution Feedback
2502.02827 (2025-02)
COFFE: A Code Efficiency Benchmark for Code Generation
2602.06953 (2026-02)
DAWN: Dependency-Aware Fast Inference for Diffusion LLMs
and 2 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Task automation has been greatly empowered by the recent advances in Large Language Models (LLMs) via Python code, where the tasks range from software engineering development to general-purpose reasoning. While current benchmarks have shown that LLMs can solve tasks using programs like human developers, the majority of their evaluations are limited to short and self-contained algorithmic tasks or standalone function calls. Solving challenging and practical tasks requires the capability of utilizing diverse function calls as tools to efficiently implement functionalities like data analysis and web development. In addition, using multiple tools to solve a task needs compositional reasoning by accurately understanding complex instructions. Fulfilling both of these characteristics can pose a great challenge for LLMs. To assess how well LLMs can solve challenging and practical tasks via programs, we introduce BigCodeBench, a benchmark that challenges LLMs to invoke multiple function calls as tools from 139 libraries and 7 domains for 1,140 fine-grained tasks. To evaluate LLMs rigorously, each task encompasses 5.6 test cases with an average branch coverage of 99%. In addition, we propose a naturallanguage-oriented variant of BigCodeBench, BigCodeBench-Instruct, that automatically transforms the original docstrings into short instructions only with essential information. Our extensive evaluation of 60 LLMs shows that LLMs are not yet capable of following complex instructions to use function calls precisely, with scores up to 60%, significantly lower than the human performance of 97%. The results underscore the need for further advancements in this area.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2406.15877")
get_code_for_paper("2406.15877")
have("2406.15877")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2406.15877.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2406.15877.svg)](https://syntology.ai/paper/2406.15877)

Connect an agent — have() is free.