Benchmark Radar™
Star —

ComplexFuncBench

Explore score history ↗

ComplexFuncBench is a benchmark designed to evaluate large language models' capabilities in handling complex function calling scenarios. It encompasses multi-step and constrained function calling tasks that require long-parameter filling, parameter value reasoning, and managing contexts up to 128k tokens. The benchmark includes 1,000 samples across five real-world scenarios.

Modality
text
Evaluates
long context, reasoning, structured output, tool calling
Openness
unknown
Importer
LLM Stats

Source provenance

ComplexFuncBench results and reported scores

Reported benchmark scores · LLM Stats

Reported scores: 7

ModelOrganizationReported valueReportedEvidence
GPT-4.1OpenAI0.6552025-04-14evidence
GPT-4.1 miniOpenAI0.4932025-04-14evidence
GPT-4.1 nanoOpenAI0.0572025-04-14evidence
GPT-4.5OpenAI0.632025-02-27evidence
GPT-4oOpenAI0.6652024-08-06evidence
Nova 2 SonicAmazon0.6522025-12-02evidence
o3-miniOpenAI0.1762025-01-30evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite ComplexFuncBench?

Related long context, reasoning, structured output, tool calling benchmarks