The docs read /v1/models rates from its per-million keys and say that pricing.prompt and completion are per token.

apps/docs/components/models-catalog.tsx

@@ -15,7 +15,9 @@ import { PROVIDER_ICONS, providerKey } from '@/components/provider-icons';
 
 const ENDPOINT = 'https://api.hanzo.ai/v1/models';
 
-type Pricing = { input: number | null; output: number | null; cacheRead: number | null; cacheWrite: number | null };
+// `prompt`/`completion` are USD per token (OpenRouter's keys); the per-million
+// rates sit under keys that name the unit.
+type Pricing = { prompt?: string; completion?: string; input_per_million?: number; output_per_million?: number };
 type Model = {
   id: string;
   name: string;
@@ -41,7 +43,7 @@ const FAMILY_ICON: Record<string, typeof Sparkle> = { Sparkle, Zap, Cpu, Box };
 // catalogue's raw key; these are the names we use for them.
 const OURS: Record<string, string> = { hanzo: 'Hanzo', zenlm: 'Zen' };
 
-// Prices come back as USD per 1M tokens.
+// Prices shown are USD per 1M tokens.
 function price(v: number | null | undefined): string {
   if (v == null) return '—';
   if (v === 0) return 'Free';
@@ -148,8 +150,8 @@ function ModelRow({ m }: { m: Model }) {
         </YStack>
       </Td>
       <Td align="right">{ctx(m.context_window)}</Td>
-      <Td align="right">{price(p?.input)}</Td>
-      <Td align="right">{price(p?.output)}</Td>
+      <Td align="right">{price(p?.input_per_million)}</Td>
+      <Td align="right">{price(p?.output_per_million)}</Td>
       <Td align="right">
         <XStack gap={4} flexWrap="wrap" justify="flex-end">
           {capabilities(m).slice(0, 4).map((f) => (

apps/docs/content/docs/decisions.mdx

@@ -195,8 +195,9 @@ and each call is a row in the org's usage (`GET /v1/billing/usage`, model
 | `typesafe/jev-1.13`, `~typesafe/jev-latest` | $0.03 per call |
 
 Kai's rate is its row in the model catalogue every model is billed from:
-`GET https://api.hanzo.ai/v1/models` lists `kai` with `pricing.input` 0.021 and
-`pricing.output` 0. A refused or failed call is not billed.
+`GET https://api.hanzo.ai/v1/models` lists `kai` with `pricing.input_per_million`
+0.021 (`pricing.prompt` `"0.000000021"` per token) and `pricing.output_per_million`
+0. A refused or failed call is not billed.
 
 ## Errors
 

apps/docs/content/docs/guides/migrate/jev.mdx

@@ -34,8 +34,9 @@ curl -sS https://api.hanzo.ai/v1/models | jq '.data[] | select(.id == "kai")'
 
 Step 2 needs an org whose plan covers paid models; without one it answers
 `402` with `"code": "plan_required"` and a link to pick a plan. Step 3 answers
-`"outputs": ["decision"]` and `"pricing": {"input": 0.021, "output": 0}`: the
-dollars per million input tokens every `kai` call is billed at.
+`"outputs": ["decision"]` and `"pricing"` with `"input_per_million": 0.021` and
+`"output_per_million": 0`: the dollars per million input tokens every `kai` call
+is billed at (`"prompt": "0.000000021"` is the same rate per token).
 
 ## What changes
 

apps/docs/content/docs/models/enso.mdx

@@ -103,7 +103,7 @@ The managed ids, from the gateway:
 
 ```bash
 curl -s https://api.hanzo.ai/v1/models \
-  | jq -r '.data[] | select(.id | startswith("enso")) | "\(.id)\t\(.context_window)\t\(.pricing.input)/\(.pricing.output)"'
+  | jq -r '.data[] | select(.id | startswith("enso")) | "\(.id)\t\(.context_window)\t\(.pricing.input_per_million)/\(.pricing.output_per_million)"'
 ```
 
 | Id | Reasoning | Note |

apps/docs/content/docs/models/index.mdx

@@ -54,14 +54,16 @@ needs no key:
 
 ```bash
 curl -s https://api.hanzo.ai/v1/models \
-  | jq -r '.data[] | select(.id | test("^(zen|enso)")) | "\(.id)\t\(.owned_by)\t\(.pricing.input)/\(.pricing.output)"'
+  | jq -r '.data[] | select(.id | test("^(zen|enso)")) | "\(.id)\t\(.owned_by)\t\(.pricing.input_per_million)/\(.pricing.output_per_million)"'
 ```
 
 <ModelsCatalog />
 
 ## What a rate means
 
-Rates are per million tokens, input and output priced separately, and they are
+Rates are per million tokens (`pricing.input_per_million`, `output_per_million`;
+`pricing.prompt` and `completion` state the same rates per token, as OpenRouter
+does), input and output priced separately, and they are
 the same whether you call from the CLI, an SDK, HTTP or an MCP tool — the price
 is a property of the model, not of the surface you reach it through.
 

apps/docs/content/docs/models/zen.mdx

@@ -75,7 +75,7 @@ needs no key:
 
 ```bash
 curl -s https://api.hanzo.ai/v1/models \
-  | jq -r '.data[] | select(.owned_by == "zenlm") | "\(.id)\t\(.context_window)\t\(.pricing.input)/\(.pricing.output)"'
+  | jq -r '.data[] | select(.owned_by == "zenlm") | "\(.id)\t\(.context_window)\t\(.pricing.input_per_million)/\(.pricing.output_per_million)"'
 ```
 
 The gateway is the source of truth for ids, windows and rates. If this page and

apps/docs/scripts/gen-pricing-page.ts

@@ -177,12 +177,19 @@ async function fetchCatalogue(): Promise<Map<string, Rate & { context: number |
       console.warn(`[pricing] catalogue answered ${r.status}`);
       return out;
     }
+    // `pricing.prompt`/`completion` are USD per token; the per-million rates this
+    // page prints sit under `input_per_million`/`output_per_million`.
     const d = (await r.json()) as {
-      data?: Array<{ id: string; context_window?: number | null; pricing?: Rate }>;
+      data?: Array<{
+        id: string;
+        context_window?: number | null;
+        pricing?: { input_per_million?: number; output_per_million?: number };
+      }>;
     };
     for (const m of d.data ?? []) {
-      if (m.pricing?.input == null) continue;
-      out.set(m.id, { ...m.pricing, context: m.context_window ?? null });
+      const input = m.pricing?.input_per_million;
+      if (input == null) continue;
+      out.set(m.id, { input, output: m.pricing?.output_per_million ?? null, context: m.context_window ?? null });
     }
   } catch (e) {
     console.warn(`[pricing] catalogue unreachable: ${(e as Error).message}`);