reasoning benchmark
COLLIE
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.
Updated Aug 7, 2026
Published models10
Registry coverage10
MetricScore
EvidenceB
COLLIE leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | OP | 99.0% | 100.0% | 10 | C | |
| 2 | OP | 98.7% | 88.9% | 10 | C | |
| 3 | OP | 98.4% | 77.8% | 10 | C | |
| 4 | MA | 95.8% | 66.7% | 10 | C | |
| 5 | OP | 72.3% | 55.6% | 10 | C | |
| 6 | OP | 65.8% | 44.4% | 10 | C | |
| 7 | MA | 62.9% | 33.3% | 10 | C | |
| 8 | OP | 61.0% | 22.2% | 10 | C | |
| 9 | OP | 54.6% | 11.1% | 10 | C | |
| 10 | OP | 42.5% | 0.0% | 10 | C |
Score distribution
Top published rows on the benchmark's original scale.
COLLIE highlights
The top published results on this benchmark's own scale.
What is COLLIE?
Definition and scoring fields from the benchmark registry.
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.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- COLLIE
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- collie|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about COLLIE.
Which model scores highest on COLLIE?
GPT-5 is currently ranked first with 99.0%.
What does COLLIE measure?
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.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
10 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is marked as eligible for the current LLMBoard capability methodology.