Models

A registry that records where its numbers came from

Every price carries a source. Every quality tier carries a source. A model with no price on file is infeasible under a cost ceiling — never free.

ModelTierContextMeasured behaviour
claude-opus-5premium200KThinking by default. Measured p90 output is materially higher than non-thinking peers, which the planner accounts for.
claude-sonnet-5balanced200KScored 32/32 on our benchmark at 2.6× less than Opus — the strongest single-model baseline we test against.
claude-haiku-4-5economy200KFailed exactly 5 of 32 cases, all in the hard band: multi-step arithmetic and multi-hop reasoning.

Anthropic models are available today. OpenAI adapters are implemented but ship without pricing rows, so those endpoints fail closed under a cost ceiling until a price with a recorded source is added.

Tiers are priors, not facts

Wide uncertainty by design

An unmeasured route carries deliberately large uncertainty, which pushes its lower-confidence bound down. It therefore cannot satisfy a high quality floor on optimism alone — the router escalates or refuses instead.

Evidence can overturn a tier

Graded trials update a posterior seeded by the tier. If measurements contradict the vendor’s claim, the measurements win — otherwise the tier would be unfalsifiable.

Output length is measured, not assumed

Thinking models bill reasoning as output. We use measured output tokens per model and task type, and refuse a candidate whose measured p90 exceeds your token budget rather than returning an empty answer.

Add your own endpoints

The registry is yours to extend — with a price source recorded against each.

Get started