Velo Model Strategy
Three strategies. One intelligent routing engine.
Velo is not tied to one model for every job. Its routing engine can evaluate eligible AI models against the capabilities required for the task, expected performance, and AI cost.
Cost Optimized
Maximize AI efficiency.
Prioritizes cost efficiency while still requiring the model capabilities needed to complete the task.
60% cost efficiency · 20% balanced fit · 20% performance
Balance capability and efficiency.
Velo's balanced strategy weighs model fit, performance, and AI cost to select an appropriate model for the work.
20% cost efficiency · 60% balanced fit · 20% performance
Performance
Prioritize maximum capability.
Weights model performance most heavily for workloads where capability matters more than AI cost.
10% cost efficiency · 20% balanced fit · 70% performance
Velo chooses the model. Your development workflow stays the same.
Files, diffs, Versions, terminal, runtime, preview, Git, and developer controls remain around the work regardless of the model selected by the routing layer.