Decision Index (public suite)

Decision Index (public suite) is a benchmark for the abilities the benchmark was built to measure. The official description says: "Public decision-suite benchmark (v0.3 public suite) in multimodalart's jev-decision-index leaderboard on Hugging Face; measures binary/multi-choice decision quality". Below we describe in simple words what benchmarks of this type generally examine.

Decision Index (public suite) measures the abilities the benchmark was built to measure. In detail, it checks:

• tasks representative of the benchmark's stated purpose
• consistent scoring rules so models can be compared fairly
• difficulty calibrated so top models do not max out easily

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: Public decision-suite benchmark (v0.3 public suite) in multimodalart's jev-decision-index leaderboard on Hugging Face; measures binary/multi-choice decision quality. Values reported as percentage. Source: https://huggingface.co/spaces/multimodalart/jev-decision-index

What is the Decision Index (public suite) benchmark?

Decision Index (public suite) is a benchmark for the abilities the benchmark was built to measure. The official description says: 'Public decision-suite benchmark (v0.3 public suite) in multimodalart's jev-decision-index leaderboard on Hugging Face; measures binary/multi-choice decision quality'. Below we describe in simple words what benchmarks of this type generally examine.

What does Decision Index (public suite) test?

Decision Index (public suite) mainly measures the abilities the benchmark was built to measure. Concretely, it checks: tasks representative of the benchmark's stated purpose; consistent scoring rules so models can be compared fairly; difficulty calibrated so top models do not max out easily.

How is the Decision Index (public suite) 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 Decision Index (public suite) data and scores?

The official source is https://huggingface.co/spaces/multimodalart/jev-decision-index. 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/decision-index/scores) or pull it directly from the Open LLM Radar API with any HTTP client.