Model Pricing API

curl --request GET \
  --url https://api.apimart.ai/api/pricing/model

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.apimart.ai/api/pricing/model",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.apimart.ai/api/pricing/model"

	req, _ := http.NewRequest("GET", url, nil)

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.apimart.ai/api/pricing/model")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)

response = http.request(request)
puts response.read_body

GET

api

pricing

model

Model Pricing API

curl --request GET \
  --url https://api.apimart.ai/api/pricing/model

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.apimart.ai/api/pricing/model",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "GET",
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.apimart.ai/api/pricing/model"

	req, _ := http.NewRequest("GET", url, nil)

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

require 'uri'
require 'net/http'

url = URI("https://api.apimart.ai/api/pricing/model")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Get.new(url)

response = http.request(request)
puts response.read_body

Examples use qwen3.8-max, but the data.pricing shape is shared by all TokenPricingV2 models.

GET /api/pricing/model?model=qwen3.8-max

Model behavior and billing rules: qwen3.8-max guide.

This endpoint does not require authentication. Do not send Authorization.

Request parameters

string

required

Model ID, e.g. qwen3.8-max. Required; omitting it returns 400 Missing model parameter.

Read only `data.pricing`

The response may include several overlapping price blocks — frontends should only use data.pricing:

Block	Purpose	Frontend
`data.pricing`	Flat, self-describing display view	✅ Use this
`data.token_price`	Legacy flat copy	❌ Compatibility only; missing fields
`data.token_pricing`	Internal snapshot	❌ Implementation details

token_price may omit fields like explicit_cached_input, so estimates run high. New billing dimensions are added only under pricing.

{
  "token_price": {
    "input": 1.714286,
    "cached_input": 0.214286,
    "cache_write": 2.142857,
    "output": 5.142857
  },
  "pricing": {
    "rates": {
      "input": 1.714286,
      "cached_input": 0.214286,
      "cache_write": 2.142857,
      "explicit_cached_input": 0.142857,
      "output": 5.142857
    }
  }
}

`rates` vs `effective_rates`

Field	Meaning
`rates`	List / strikethrough price
`effective_rates`	What to display as payable (discounts applied)
`discount_rate`	Model discount (`1` = none)
`group_ratio`	Group multiplier
`price_factor`	`discount_rate × group_ratio`

{
  "discount_rate": 1,
  "group_ratio": 0.8,
  "price_factor": 0.8,
  "rates": { "input": 1.714286, "output": 5.142857 },
  "effective_rates": { "input": 1.3714288, "output": 4.1142856 }
}

Use effective_rates for UI prices — do not multiply yourself. Use rates only when showing list price alongside.

group is often default: public pricing pages quote default. A user’s real charge uses their group and may be lower than the displayed price.

rates / effective_rates use usd_per_million_tokens (USD per 1M tokens).

Tool prices: `extras.tools` is list price only

Asymmetry in the current API:

	List	Payable
Tokens	`rates`	`effective_rates` (precomputed)
Built-in tools	`extras.tools[x].price`	No effective field — multiply by `price_factor` yourself

const listed = pricing.extras.tools.web_search.price;
const actual = listed * pricing.price_factor; // display this

Respect unit — do not hardcode “per 1k calls”:

`unit`	Meaning	Cost
`usd_per_1000_queries`	Per 1000 calls	`price / 1000 × count`
`usd_per_page`	Per page	`price × pages` (do not divide by 1000)

Three-state field semantics

Case	JSON	Meaning
Present	`"explicit_cached_input": 0.142857`	Bill at this rate
Key missing	—	Dimension not applicable (not zero, not free)
Explicit `0`	`"price": 0`	Actually free — show as free

Examples:

Missing output_thinking → single output rate; do not render “thinking is free”. For qwen3.8-max, thinking cannot be disabled and there is one output price.
Missing extras.google_web_search → Vertex-only; Bailian-style search is under extras.tools.web_search.
Tool with price: 0 → explicit free (e.g. promo); show free.

Multi-tier: only `tier_count`

{
  "billing_type": "tiered_token",
  "tier_count": 1
}

Check	Behavior
`tier_count === 1`	Do not render a tier table; use `rates` / `effective_rates`
`tier_count > 1`	`tiers[]` present; `up_to_input_tokens` is inclusive upper bound; missing key = uncapped (usually last tier)

Tier rule: pick one tier from total input tokens of the request, then price the whole request at that tier (not progressive). Values in tiers are list prices — multiply by price_factor.

Do not use billing_type === "tiered_token" to detect multi-tier pricing — single-tier models can still use that value. Only trust tier_count.

`limits` (not prices)

{
  "limits": {
    "max_input_tokens": 983616,
    "max_output_tokens": 131072,
    "supports_cache_read": true,
    "supports_cache_write": true
  }
}

supports_cache_*: capability flags; if false, do not present cache rates as available even if present in rates.
max_output_tokens: pre-charge style cap when max_tokens is omitted — not a price.

TypeScript types

type ToolItem = {
  unit: "usd_per_1000_queries" | "usd_per_page";
  price: number; // list price; × price_factor for display
  precharge_queries?: number;
};

type Rates = {
  input?: number;
  cached_input?: number;
  explicit_cached_input?: number;
  cache_write?: number;
  cache_write_5m?: number;
  cache_write_1h?: number;
  output?: number;
  output_thinking?: number;
  text_input?: number;
  cached_text_input?: number;
  image_input?: number;
  cached_image_input?: number;
  text_output?: number;
  image_output?: number;
};

type ModelPricing = {
  billing_type: string;
  source: "v1" | "v2";
  unit: "usd_per_million_tokens";
  pricing_mode: "standard" | "image_modalities";
  discount_rate: number;
  group: string;
  group_ratio: number;
  price_factor: number;
  resolved_from?: string;
  rates?: Rates;
  effective_rates?: Rates;
  tier_count: number;
  tiers?: (Rates & { up_to_input_tokens?: number })[];
  limits?: {
    max_input_tokens?: number;
    max_output_tokens?: number;
    supports_cache_read: boolean;
    supports_cache_write: boolean;
  };
  extras?: {
    tools?: Record<string, ToolItem>;
    google_web_search?: ToolItem;
  };
};

Tool row helper

const toolRows = Object.entries(pricing.extras?.tools ?? {}).map(
  ([tool, item]) => ({
    tool,
    actual: item.price * pricing.price_factor,
    suffix: item.unit === "usd_per_page" ? "/ page" : "/ 1k calls",
    free: item.price === 0,
  }),
);

Common mistakes

Wrong	Right
Read `token_price`	Read `pricing.rates` / `pricing.effective_rates`
Multiply `rates` yourself	Use `effective_rates`
Show raw `extras.tools[x].price`	Multiply by `price_factor`
Hardcode “per 1k calls”	Branch on `item.unit`
Treat missing keys as 0 / free	Missing = N/A; only explicit `0` is free
Use `billing_type` for tiers	Use `tier_count > 1`
Read `tiers` for single-tier	Use `rates`
Mix `pricing_mode` fields	`standard` vs `image_modalities` are exclusive

Request examples

curl --request GET \
  --url 'https://api.apimart.ai/api/pricing/model?model=qwen3.8-max' \
  --header 'Accept: application/json'

import requests

url = "https://api.apimart.ai/api/pricing/model"
params = {"model": "qwen3.8-max"}
headers = {"Accept": "application/json"}
print(requests.get(url, params=params, headers=headers).json())

Overview

Models

Text Series

Image Series

Video Series

Audio Series

Moderation Series

Upload Management

Task Management

Account Management

Model Pricing API

Request parameters

Read only `data.pricing`

`rates` vs `effective_rates`

Tool prices: `extras.tools` is list price only

Three-state field semantics

Multi-tier: only `tier_count`

`limits` (not prices)

TypeScript types

Tool row helper

Common mistakes

Request examples

​Request parameters

​Read only data.pricing

​rates vs effective_rates

​Tool prices: extras.tools is list price only

​Three-state field semantics

​Multi-tier: only tier_count

​limits (not prices)

​TypeScript types

​Tool row helper

​Common mistakes

​Request examples

Request parameters

Read only `data.pricing`

`rates` vs `effective_rates`

Tool prices: `extras.tools` is list price only

Three-state field semantics

Multi-tier: only `tier_count`

`limits` (not prices)

TypeScript types

Tool row helper

Common mistakes

Request examples