Classification Benchmark

CLINC150+OOS

CLINC150 with out-of-scope (OOS) intent classification, scored as macro-F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Category

Classification

Score type

percent

Scale

100%

higher is better

Models tested

2

What is CLINC150+OOS?

CLINC150+OOS is a benchmark for sorting inputs into the correct categories. The official description says: "CLINC150 with out-of-scope (OOS) intent classification, scored as macro-F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite". Below we describe in simple words what benchmarks of this type generally examine.

What does it test?

CLINC150+OOS measures sorting inputs into the correct categories. In detail, it checks:

  • choosing the right category from a fixed set of options
  • handling out-of-scope inputs that belong to no category
  • consistent labels across many short inputs
  • robustness to paraphrasing and noise

How to read the score

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: CLINC150 with out-of-scope (OOS) intent classification, scored as macro-F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Frequently asked questions

What is the CLINC150+OOS benchmark?

CLINC150+OOS is a benchmark for sorting inputs into the correct categories. The official description says: 'CLINC150 with out-of-scope (OOS) intent classification, scored as macro-F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite'. Below we describe in simple words what benchmarks of this type generally examine.

What does CLINC150+OOS test?

CLINC150+OOS mainly measures sorting inputs into the correct categories. Concretely, it checks: choosing the right category from a fixed set of options; handling out-of-scope inputs that belong to no category; consistent labels across many short inputs.

How is the CLINC150+OOS score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get CLINC150+OOS data and scores?

The official source is https://clef-evals.workers-ai-mle.workers.dev/. The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/clinc150-oos/scores) or pull it directly from the Open LLM Radar API with any HTTP client.

Model scores 2 models tested

Reset
# Model Score Date Source Verified
1 Clef
Cloudflare
97.4%
— Report Verified
2 Clef-Flash
Cloudflare
66.8%
— Report Verified

What now?

You've seen how models score on CLINC150+OOS — here's what to do next.