Real tasks. Measured evidence.

Choose a model with your own tasks.

Compare quality, speed and cost. Inspect the evidence, then rerun the same task yourself.

Start a comparison →View test reports

Prepare a three-model comparison without signing up. Use trial quota or your own API keys.

How a comparison works · demo

01 · Complete the task✓
02 · Wait for the answer→
03 · Inspect the cost¥

Quality is marked unassessed until an explicit task check passes.

Latest published evidence

All reports →

2026-09-27 · 15 attempts · hkg1 · Retest pending

Space Bunny Alpha

JSON extraction, instruction constraints and grounded QA — pass counts, failure cases and original outputs.

Inspect results →
01

Define success

Use a versioned check or inspect the original output.

02

Compare equal conditions

Keep provider, region, settings and measurement window explicit.

03

Keep the evidence

Export results and return to the same report version.