Benchmark

COLLIE

language reasoning writing text

COLLIE is a grammar-based framework for systematic construction of constrained text generation tasks. It allows specification of rich, compositional constraints across diverse generation levels and modeling challenges including language understanding, logical reasoning, and semantic planning. The COLLIE-v1 dataset contains 2,080 instances across 13 constraint structures.

语言EN
满分1
参评模型8

模型排名

名次 模型 机构 分数 来源
1 GPT-5 OpenAI 99.0 来源 ↗
2 o3-mini OpenAI 98.7 来源 ↗
3 o3 OpenAI 98.4 来源 ↗
4 GPT-4.5 OpenAI 72.3 来源 ↗
5 GPT-4.1 OpenAI 65.8 来源 ↗
6 GPT-4o OpenAI 61.0 来源 ↗
7 GPT-4.1 mini OpenAI 54.6 来源 ↗
8 GPT-4.1 nano OpenAI 42.5 来源 ↗