discount-token Guide

Leading AI Model API Service — All-in-one setup tutorial

Overview

discount-token provides API access to leading AI models including Zhipu GLM, Volcengine, Claude, GPT, and DeepSeek, compatible with major coding tools.

Base URL

Unified Endpoint

https://discount-token.com/v1

Compatible

OpenAI Format

Simply replace Base URL and API Key in any OpenAI-compatible client.

Limits

Rate Limits

20 RPM per user. 5-hour / daily / weekly quota limits apply.

Available Models

Claude Series

claude-opus-4-8Ultimate flagship

claude-opus-4-7High performance

claude-sonnet-5Best value

claude-sonnet-4-6Balanced

claude-sonnet-4-5Stable

claude-haiku-4-5Fast & light

claude-fable-5Creative writing

GPT Series

gpt-5.6Latest flagship

gpt-5.5High performance

gpt-5.4Balanced

DeepSeek Series

deepseek-r1Deep reasoning

deepseek-v4-proPro edition

deepseek-v4-flashUltra fast

deepseek-v3Stable

Zhipu GLM Series

glm-5.2Flagship reasoning

glm-5.1High performance

glm-5Standard

glm-4.7Lightweight & fast

glm-4.6Stable

Volcengine Series

doubao-seed-2.0-proDoubao flagship

doubao-seed-2.0-codeCode specialist

doubao-seed-2.0-liteLightweight

kimi-k2.6Kimi latest

Note: Model names are all lowercase with hyphens. Typing the wrong name returns 404 Model Not Found.

Create API Key

Register / Login

Visit https://discount-token.com, sign up or log in.

Top Up

Go to the "Top Up" page, choose an amount, and pay with Alipay. Balance is credited automatically.

Create API Key

Navigate to "API Keys", click "New", and copy the generated key (format sk-xxx). Save it securely — it's only shown once.

Cursor Setup

Add Custom Model

Open Cursor → Settings → Models → Add Custom Model

Enter model name: glm-5.2

Configure OpenAI API

Field	Value
OpenAI API Key	`sk-your-key`
Base URL	`https://discount-token.com/v1`

Click Verify — green checkmark means success.

Start Using

Select model glm-5.2 in Chat or Composer.

Claude Code (Terminal)

Option 1: Environment Variables

BASH TerminalEnv vars

export OPENAI_API_KEY="sk-your-key"
export OPENAI_BASE_URL="https://discount-token.com/v1"

# Launch Claude Code
claude --model glm-5.2

Option 2: Config File

Edit ~/.claude/config.json:

JSON ~/.claude/config.jsonConfig

{
  "api_key": "sk-your-key",
  "base_url": "https://discount-token.com/v1",
  "model": "glm-5.2"
}

Codex / Codex++

Open Provider Settings

In Codex++, go to settings and find "Provider Management".

Add Provider

Add a new provider. Name: discount-token, Type: OpenAI Compatible.

Fill Configuration

Field	Value
Provider Name	discount-token
Base URL	`https://discount-token.com/v1`
API Key	`sk-your-key`
Model	`glm-5.2` or `doubao-seed-2.0-pro`

Cline / Roo Code (VS Code)

Field	Value
API Provider	OpenAI Compatible
Base URL	`https://discount-token.com/v1`
API Key	`sk-your-key`
Model	`glm-5.2`

Kilo Code

Field	Value
Provider	OpenAI Compatible
Base URL	`https://discount-token.com/v1`
API Key	`sk-your-key`
Model	`glm-5.2`

OpenCode / Crush

Edit config file opencode.json:

JSON opencode.jsonConfig

{
  "provider": "openai",
  "api_key": "sk-your-key",
  "base_url": "https://discount-token.com/v1",
  "model": "glm-5.2"
}

API Format

This service is OpenAI-compatible, supporting /v1/chat/completions and /v1/responses.

Auth

Bearer Token

Header: Authorization: Bearer <your-key>

Compatible

OpenAI SDK

Works with all OpenAI SDKs and compatible clients.

cURL

POST /v1/chat/completionsChat example

curl https://discount-token.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-key" \
  -d '{
    "model": "glm-5.2",
    "messages": [
      { "role": "user", "content": "Hello, write a short product intro" }
    ]
  }'

Python

PY Python SDKopenai ≥ 1.0

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-key",
    base_url="https://discount-token.com/v1"
)

response = client.chat.completions.create(
    model="glm-5.2",
    messages=[{"role": "user", "content": "Hello"}]
)

print(response.choices[0].message.content)

GLM-5.2 Peak Hours Notice

Important: Zhipu applies the following quota consumption rates for GLM-5.2:

Period	Time (Beijing)	Multiplier
Peak	14:00 - 18:00 daily	3x
Off-peak	18:00 - 14:00 next day	2x
Limited promo (until Sep)	Off-peak only	1x

Tips to save quota:

Use GLM-4.7 (1x) during peak hours — fast and sufficient for most tasks
Switch to GLM-5.2 for complex tasks during off-peak hours
Before Sep 30, GLM-5.2 is only 1x during off-peak — take advantage!

FAQ

Error 429 Too Many Requests

You exceeded 20 requests/minute. Wait 1 minute and retry.

Error 401 Unauthorized

Check your API Key is correct with no extra spaces.

Error 404 Model Not Found

Check model name spelling: glm-5.2 (lowercase, hyphenated).

5-hour quota exhausted

Quota resets automatically after 5 hours. Use off-peak for non-urgent tasks.

What Base URL to use?

Use https://discount-token.com/v1. If your tool auto-appends /v1, use https://discount-token.com.

Is streaming supported?

Yes. Add "stream": true to your request, identical to OpenAI.