General Capabilities Benchmark

ACOS

ACOS: aspect-category-opinion-sentiment quad classification over reviews, scored as per-review F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Category

General Capabilities

Score type

percent

Scale

100%

higher is better

Models tested

2

What is ACOS?

ACOS is a benchmark for broad everyday usefulness across domains. The official description says: "ACOS: aspect-category-opinion-sentiment quad classification over reviews, scored as per-review 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?

ACOS measures broad everyday usefulness across domains. In detail, it checks:

  • realistic, open-ended requests from varied professional fields
  • correct, well-structured and helpful responses
  • consistent quality across very different task types
  • judgement and advice quality, not just factual recall

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: ACOS: aspect-category-opinion-sentiment quad classification over reviews, scored as per-review F1 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Frequently asked questions

What is the ACOS benchmark?

ACOS is a benchmark for broad everyday usefulness across domains. The official description says: 'ACOS: aspect-category-opinion-sentiment quad classification over reviews, scored as per-review 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 ACOS test?

ACOS mainly measures broad everyday usefulness across domains. Concretely, it checks: realistic, open-ended requests from varied professional fields; correct, well-structured and helpful responses; consistent quality across very different task types.

How is the ACOS 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 ACOS 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/acos/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
33.3%
— Report Verified
2 Clef-Flash
Cloudflare
25.9%
— Report Verified

What now?

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