grok · 4 arms · 4 effortsgrok-4.5

One model across every effort and arm. Each cell is a run — open it for the full card. Failure chips mark runs with no valid slot.

effort strip

arm \ effort	none	low	medium	high
api	open run ↗ 32.1K tokens8 runs · all valid	open run ↗ 32.4K tokens8 runs · all valid	open run ↗ 34.4K tokens8 runs · all valid	open run ↗ 33.3K tokens8 runs · all valid
go	open run ↗ 18.1K tokens4 runs · all valid	—	—	—
grok-cli	—	open run ↗ — tokens4 runs · all valid	open run ↗ — tokens4 runs · all valid	open run ↗ — tokens4 runs · all valid
opencode	open run ↗ — tokens3 of 4 valid	—	—	—

Output completion tokens per effort, per arm. A read on how much each model spends as effort rises — not a quality axis.

APIGo (native)grok (CLI)opencode