Claude Code
The v2 rules inside one identical sandbox for every tool, with a provided headless Chromium for the agent’s own localhost preview. Runs are compared only within one protocol. How the protocols differ
workflow-subscription-v3 is the newest benchmark protocol, and none of its Runs is published for this model yet. Results of another protocol are not shown in their place, because Runs are compared only within one protocol. Pick another protocol above, or read how the protocols differ.