Agentic Benchmark

BRIGHT

BRIGHT evaluates reality-grounded retrieval for reasoning-heavy queries, scored as nDCG@10 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Category

Agentic

Score type

percent

Scale

100%

higher is better

Models tested

2

What is BRIGHT?

BRIGHT is a benchmark for agentic abilities such as tool use, retrieval and autonomous workflows. The official description says: "BRIGHT evaluates reality-grounded retrieval for reasoning-heavy queries, scored as nDCG@10 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?

BRIGHT measures agentic abilities such as tool use, retrieval and autonomous workflows. In detail, it checks:

  • choosing and calling the right tools for a task
  • keeping track of what it has already done and what is left
  • recovering from failed or contradicting tool results
  • completing a defined workflow end-to-end

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: BRIGHT evaluates reality-grounded retrieval for reasoning-heavy queries, scored as nDCG@10 on a 0-100 scale in Cloudflare's internal Decision Index 0.2.1 suite.

Frequently asked questions

What is the BRIGHT benchmark?

BRIGHT is a benchmark for agentic abilities such as tool use, retrieval and autonomous workflows. The official description says: 'BRIGHT evaluates reality-grounded retrieval for reasoning-heavy queries, scored as nDCG@10 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 BRIGHT test?

BRIGHT mainly measures agentic abilities such as tool use, retrieval and autonomous workflows. Concretely, it checks: choosing and calling the right tools for a task; keeping track of what it has already done and what is left; recovering from failed or contradicting tool results.

How is the BRIGHT 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 BRIGHT 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/bright/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
45.9%
— Report Verified
2 Clef-Flash
Cloudflare
39.3%
— Report Verified

What now?

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