Unified LLM gatewaybuilt for production

Enterprise customers get extra discounts — contact sales for an exclusive enterprise quote.

Pricing at a glance

Input / output · USD / 1M tokens
99.98%Service availability SLA
5,000 QPMPeak concurrency per customer
95%+Guaranteed cache hit rate
10,000+Businesses onboarded

Enterprise pricing

Prices in USD / 1M tokens. Contact sales to enable Azure, AWS, or Vertex resources.

GPT 6.1 Sol1M contextTiered context pricing
Standard77.7% off
List price $2.0 / $10.0 / $0.1
Input
$0.4
Output
$2.2
Cache
$0.0
Azure10.0% off
List price $2.0 / $10.0 / $0.1
Input
$1.8
Output
$9.0
Cache
$0.1
GPT-6 Astra1M contextTiered context pricing
Standard77.7% off
List price $10.0 / $50.0 / $1.0
Input
$2.2
Output
$11.1
Cache
$0.2
Azure10.0% off
List price $10.0 / $50.0 / $1.0
Input
$9.0
Output
$45.0
Cache
$0.9
Claude Opus 5.51M context
Standard69.3% off
List price $4.0 / $20.0 / $0.2
Input
$1.2
Output
$6.1
Cache
$0.1
AWS10.0% off
List price $4.0 / $20.0 / $0.2
Input
$3.6
Output
$18.0
Cache
$0.2
Claude Fable 5.21M context
Standard68.2% off
List price $8.0 / $40.0 / $0.2
Input
$2.5
Output
$12.7
Cache
$0.1
AWS10.0% off
List price $8.0 / $40.0 / $0.2
Input
$7.2
Output
$36.0
Cache
$0.2
Claude Sonnet 5.51M context
Standard69.3% off
List price $2.0 / $10.0 / $0.1
Input
$0.6
Output
$3.1
Cache
$0.0
AWS10.0% off
List price $2.0 / $10.0 / $0.1
Input
$1.8
Output
$9.0
Cache
$0.1
Gemini 4 Carbon1M context
Standard69.3% off
List price $2.0 / $10.0 / $0.1
Input
$0.6
Output
$3.1
Cache
$0.0
Vertex10.0% off
List price $2.0 / $10.0 / $0.1
Input
$1.8
Output
$9.0
Cache
$0.1
Deepseek V4.1 Flash 10171M context
Volcengine9.9% off
List price Peak $0.3 / $1.2 / $0.0Off-peak $0.2 / $0.6 / $0.0
Input
Peak $0.3Off-peak $0.1
Output
Peak $1.1Off-peak $0.5
Cache
Peak $0.0Off-peak $0.0
GPT Image 2.6 SunburstImage model
StandardPer image (USD / image)
  • 1K$0.2$0.089.4% off
  • 2K$0.2$0.089.4% off
  • 4K$0.2$0.169.3% off
AzurePer token (USD / 1M tokens)
  • Text input$5.0$4.510.0% off
  • Image input$8.0$7.210.0% off
  • Output$30.0$27.010.0% off
  • Cached text input$1.3$1.110.0% off
  • Cached image input$2.0$1.810.0% off

Updated 2026-11-16 (Beijing time). Prices in USD / 1M tokens. Discount (% off) = 1 − API Token Router price ÷ list price. Prices are fixed by resource type and do not move with upstream costs; DeepSeek is quoted separately at its official peak and off-peak prices. Contact sales to enable Azure, AWS, or Vertex resources.

The more you use, the less you pay

Prices above are standard enterprise rates. Enterprise customers get extra discounts based on usage — contact sales for an exclusive enterprise quote.

An AI gateway built for enterprise scale

30T+Monthly Tokens
100K+Global Users
60+Models
10K+Enterprises

One platform for every enterprise team

Give builders, finance and administrators one place to manage access, budgets and usage.

Developers & engineering

One key for many models

  • One API for 60+ models
  • Routing and API keys
  • Desktop client setup
Finance & operations

Account for every request

  • Organization and member usage
  • Member spending limits
  • Costs by model
Security & risk

Access and spending controls

  • Organization members
  • Key limits and expiry
  • Privacy and data handling
Procurement & legal

A simpler way to work together

  • Business verification
  • Contact sales
  • Payment options

60+ models. Find the right fit for your business.

Choose a model
from openai import OpenAI client = OpenAI( base_url="https://apitokenrouter.com/v1", # the only line that changes api_key="<API_TOKEN_ROUTER_API_KEY>", ) response = client.chat.completions.create( model="claude-falbe-5-2", messages=[{"role": "user", "content": "Hello"}], ) print(response.choices[0].message.content)

Integrate once. Explore 60+ models.

Use one API, model directory and configuration entry. Choose models by workload.

Token usage

Organization budget
62%
Member budget
78%

Understand your spend. Avoid surprises.

Bring organization usage, member spending and budgets into one view for your teams.

Members

Invite
SShen YinuoPending
Invite member
CancelSend invite

Built for teams

Manage members, API keys, budgets and usage in one enterprise workspace.

Data & privacy

Model training
API Token Router does not train models on your data
Encrypted transit
Data is transferred over TLS
Data handling
Follows security and privacy policies

Security and privacy for every request

Understand how requests are handled. API Token Router does not train models on your content; requests are sent to selected providers under their policies and terms.

Choose the right models for your team

Tell us what you need and get an enterprise solution for your team.

API Token Router - Claude / GPT API 企业服务|AWS、Azure、Vertex资源报价