V*

Reported in recent model launches

These model reports include this benchmark. Their harnesses, task subsets, scoring policies, and resource budgets can differ; follow each report’s evaluation notes before joining the numbers.

What it measures

Detailed visual questions test the ability to locate and interpret relevant evidence inside an image.

How to read the result

Inspect whether the model actually sees the necessary region and resolves the question rather than relying on background expectations.

Limits and comparison traps

Image navigation, cropping and Python access change the system’s observations. Pin the visual tool configuration and task version.

Benchmark reference notes

Evaluates active visual search for fine-grained details inside high-resolution crowded images.

Where this fits in an evaluation suite

Ask what the task actually requires: recall, deduction, search, or a mixture. Separate closed-book reasoning from retrieval and tool-assisted problem solving.

Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.

A reproducible comparison

Pin the benchmark release, task split, model configuration, tools, sampling budget, and evaluation harness described by the primary source.

Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.