Transparent pricing. Choose what fits.
Every plan is billed by actual token usage and includes access to all 9 models. Default has no monthly fee, while Pro and Max offer lower input and output rates.
99 production modelsLive model and cache pricing
20%Pro saves 20% on input and outputAvailable across all 9 models
50%Max saves 50% on input and outputDesigned for high-volume production
Zero monthly feeDefault
Access every production model for $0 per month and pay only for the tokens you use. A practical fit for development, personal apps, and workloads that need flexible spend.
$0/ month
- 9 production models with usage-based billingCurrent prices shown below
- Access all 9 production models
- No subscription; top up only when needed
- No plan-level token cap
Start with Default→
20% OFF MODEL RATES
Steady productionPro
For developers and growing teams with consistent API usage. Every model costs 20% less than Default, balancing predictable savings with everyday production needs.
$5.99/ month
$1 model-usage credit included
- 20% off every input and output rateCurrent prices shown below
- 20% off every Default model rate
- Access all 9 production models
- No plan-level token cap
Choose Pro→
50% OFF · HIGH-VOLUME VALUE
Scale with confidenceMax
For high-volume applications, agents, and automation. Every model's input and output costs 50% less than Default, providing predictable savings for high-volume production.
$39.99/ month
$5 model-usage credit included
- 50% off every input and output rateCurrent prices shown below
- 50% off every Default input and output rate
- Built for high-volume production workloads
- No plan-level token cap
Choose Max→
Model rates
Official API vs. every AnRouter plan
USD per 1M tokens
Each model shows the official API reference alongside Default, Pro, and Max input, cache-read, and output rates. Savings are calculated separately for input and output.
DeepSeek official API price reference
Input$0.15
Cache read$0.003
Output$0.599
DefaultPay as you go
Input$0.0847% lower
Output$0.350% lower
Cache read $0.015
Pro20% off I/O
Input$0.06457% lower
Output$0.2460% lower
Cache read $0.015
Max50% off I/O
Input$0.0473% lower
Output$0.1575% lower
Cache read $0.015
Z.AI official API price reference
Input$1.198
Cache read$0.091
Output$4.194
DefaultPay as you go
Input$0.4265% lower
Output$1.2570% lower
Cache read $0.08
Pro20% off I/O
Input$0.33672% lower
Output$176% lower
Cache read $0.08
Max50% off I/O
Input$0.2182% lower
Output$0.62585% lower
Cache read $0.08
Z.AI official API price reference
Input$1.198
Cache read$0.1656
Output$4.194
DefaultPay as you go
Input$0.4562% lower
Output$1.369% lower
Cache read $0.08
Pro20% off I/O
Input$0.3670% lower
Output$1.0475% lower
Cache read $0.08
Max50% off I/O
Input$0.22581% lower
Output$0.6585% lower
Cache read $0.08
Z.AI official API price reference
Input$0.12
Cache read$0.015
Output$0.419
DefaultPay as you go
Input$0.04562% lower
Output$0.1662% lower
Cache read $0.012
Pro20% off I/O
Input$0.03670% lower
Output$0.12869% lower
Cache read $0.012
Max50% off I/O
Input$0.022581% lower
Output$0.0881% lower
Cache read $0.012
MiniMax official API price reference
Input$0.15
Cache read$0.015
Output$0.599
DefaultPay as you go
Input$0.1220% lower
Output$0.433% lower
Cache read $0.038
Pro20% off I/O
Input$0.09636% lower
Output$0.3247% lower
Cache read $0.038
Max50% off I/O
Input$0.0660% lower
Output$0.267% lower
Cache read $0.038
Qwen official API price reference
Input$0.449
Cache read$0.085
Output$1.797
DefaultPay as you go
Input$0.0882% lower
Output$0.478% lower
Cache read $0.025
Pro20% off I/O
Input$0.06486% lower
Output$0.3282% lower
Cache read $0.025
Max50% off I/O
Input$0.0491% lower
Output$0.289% lower
Cache read $0.025
Qwen official API price reference
Input$0.12
Cache read$0.016
Output$0.404
DefaultPay as you go
Input$0.0650% lower
Output$0.1270% lower
Cache read $0.014
Pro20% off I/O
Input$0.04860% lower
Output$0.09676% lower
Cache read $0.014
Max50% off I/O
Input$0.0375% lower
Output$0.0685% lower
Cache read $0.014
Xiaomi MiMo official API price reference
Input$0.449
Cache read$0.0037
Output$0.899
DefaultPay as you go
Input$0.2544% lower
Output$0.544% lower
Cache read $0.04
Pro20% off I/O
Input$0.255% lower
Output$0.455% lower
Cache read $0.04
Max50% off I/O
Input$0.12572% lower
Output$0.2572% lower
Cache read $0.04
Xiaomi MiMo official API price reference
Input$0.15
Cache read$0.003
Output$0.3
DefaultPay as you go
Input$0.0660% lower
Output$0.1550% lower
Cache read $0.01
Pro20% off I/O
Input$0.04868% lower
Output$0.1260% lower
Cache read $0.01
Max50% off I/O
Input$0.0380% lower
Output$0.07575% lower
Cache read $0.01
Billing details
How the plans work
Do Pro and Max include credit?
Yes. Pro includes $1 and Max includes $5 in model-usage credit. Additional usage is charged from your wallet at the active plan rate.
Is there a plan-level token limit?
No. Default, Pro, and Max do not impose a fixed token allowance. Service-level fair-use and safety controls may still protect platform availability.
When does the discount apply?
The active membership group determines the rate: Default pays standard rates, Pro receives 20% off, and Max receives 50% off.
