linecard

Perfectly configured.
Full native power.

The configuration layer every team rebuilds by hand: pricing, capabilities, and configuration for every model. Current, verified, queryable.

Browse the registry See Switchboard

Tracked per model

Pricing per tokenPrice historyContext windowTool callingVisionReasoningPrompt cachingStreamingSupported parametersSystem prompt structureCache positioning

How does Linecard help?

Every model, every capability

Track what each model actually supports: tool calling, vision, reasoning, prompt caching. Every model is auto-configured to perform with its full capability set.

See the matrix

Verified by a human

Every price and every capability is verified by a person before it goes live, and pricing is passthrough: the rate you see is the provider's rate.

View pricing

Configurations, not just prices

Tool formats, cache behavior, parameter sets: providers change them constantly. Through Linecard, every model arrives already configured to perform; you never tune a setting.

See how routing uses it

Capability checks on every request

Every request is checked against what the model can actually do before it runs. Ask for a capability the model lacks and you get a clear warning up front; a request the model cannot serve is stopped before it costs you anything, not answered with a cryptic provider error.

See the capability matrix

A price ledger, not a snapshot

Every price change is a new point in a per-model ledger. See the trend, audit any bill against the rate at the time.

View price history

Eight providers, 63 models, live

The catalog is refreshed daily. When a new model ships, it shows up ready to route: instant models, on demand.

Browse the registry

One source of truth for routing

Linecard is what Switchboard reads to route. If a model is in Linecard, it is routable, priced, and configured; there is no separate setup step.

See Switchboard

Sample rows

What a Linecard row looks like

Model	Input / 1M	Output / 1M	Cache (read / write)	Tools	Vision
claude-sonnet-5	$2.00	$10.00	0.10× / 1.25×	✓	✓
gpt-5.5	$5.00	$30.00	0.10× / —	✓	✓
kimi-k2.6	$0.95	$4.00	0.17× / —	✓	—
deepseek-v4-flash	$0.14	$0.28	0.02× / —	✓	—
mistral-large-latest	$0.50	$1.50	—	✓	—

Indicative pricing for illustration. The live Linecard reflects the current rate every provider charges, with timestamped updates.

Linecard FAQs

Is Linecard the same as a model catalog?

Similar, but Linecard is opinionated about what gets tracked: pricing, capability flags, and the configuration each model needs to perform. Catalogs list names. Linecard is structured for cost modelling and capability-based routing.

How fresh is the pricing data?

Checked against every provider daily, and every change is verified by a person before it goes live. Each rate is timestamped, so any bill can be audited against the rate at the time.

Can I query Linecard directly?

Yes. Browse and filter the registry by capability, price, and provider in the portal, or pull the same data through the read API.

Is Linecard open source?

Not yet. The data shape is documented; if open access becomes the right move for distribution, we'll consider opening the data first and the platform separately.

The other half

One API for every model in the registry

Switchboard routes to every enabled provider in Linecard. OpenAI-compatible or Anthropic native wire format, per-end-user attribution, one invoice.