o10Last updated 2026-06-09

Inference model catalog

Each model page maps tier, venue pricing, and when eval-gated routing should select it — not default frontier tiers.

Each model page maps tier, venue pricing, and when eval-gated routing should select it — with links to savings and pricing detail.

Spread observed
638×
Routing modes
shadow → enforce
Framework
KYI
Dashboards observe.
o10 enforces.

Every page in this index follows the same structure as the home site — answer-first, passage blocks, operational steps, and expanded FAQs.

Start hereQuick overview

How to use this index

What is the o10 model catalog?

Dedicated pages per model tier with gateway, committed, and open-weight pricing — plus use-case guidance and cross-venue routing notes.

Every model links to savings matrix and pricing detail pages.

Why model pages separate from pricing?

Pricing pages are provider×model venue matrices. Model pages are model-centric — tier, use cases, and cross-venue routing guidance.

Phase 2 adds programmatic model profiles for the full routing graph.

How should you use this catalog?

Pick your workload, find candidate models, verify eval equivalence in shadow mode, then enforce the cheapest compliant route.

Links to savings matrix and routing guides per use case.

FAQFrequently asked questions

Common questions

How many models are in the catalog?

Phase 2 ships 22 model profiles across frontier, sonnet, mini, reasoning, and open-weight tiers — expanding as providers add models.

Are prices guaranteed?

Figures are June 2026 survey benchmarks for comparison. Always verify against provider list prices; o10 shadow mode proves savings on your traffic.

How do models connect to savings pages?

Each model links to use-case×model savings estimates — workload volume, current rate, routed rate, and compliant savings at your quality floor.

o10Set the envelope. o10 holds it.

See what you're overpaying.

Paste a week of traffic. Get the number that books the audit.

See what you're overpaying