ZeroBench_main w/ python (pass@5)

ZeroBench_main w/ python (pass@5) is a benchmark for understanding images combined with text. The official description says: "Tools-augmented ZeroBench main set, pass@5, averaged over 5 runs". Below we describe in simple words what benchmarks of this type generally examine.

ZeroBench_main w/ python (pass@5) measures understanding images combined with text. In detail, it checks:

• answering questions about images (figures, photos, charts, memes)
• reading text inside images and connecting it to the question
• visual reasoning such as counting, locating and comparing objects
• combining what is seen with general knowledge

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: Tools-augmented ZeroBench main set, pass@5, averaged over 5 runs.

What is the ZeroBench_main w/ python (pass@5) benchmark?

ZeroBench_main w/ python (pass@5) is a benchmark for understanding images combined with text. The official description says: 'Tools-augmented ZeroBench main set, pass@5, averaged over 5 runs'. Below we describe in simple words what benchmarks of this type generally examine.

What does ZeroBench_main w/ python (pass@5) test?

ZeroBench_main w/ python (pass@5) mainly measures understanding images combined with text. Concretely, it checks: answering questions about images (figures, photos, charts, memes); reading text inside images and connecting it to the question; visual reasoning such as counting, locating and comparing objects.

How is the ZeroBench_main w/ python (pass@5) 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 ZeroBench_main w/ python (pass@5) data and scores?

The official source is https://huggingface.co/moonshotai/Kimi-K3.1-Pro. 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/zerobench-main-with-python-pass5/scores) or pull it directly from the Open LLM Radar API with any HTTP client.