NIH/Multi-needle

long context official site →

Multilingual: no

Multi-needle in a haystack benchmark for evaluating long-context comprehension capabilities of language models by testing retrieval of multiple target pieces of information from extended documents

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: long_context. Language: en. Verified by llm-stats: no.