Event Update Notes

Due to a business restructuring, effective immediately, the "Flagship Model Value Subscription" and the "Lightweight All-in-One Subscription" will be officially merged into the "Value Resource Package," and their pricing will be standardized. If you have previously purchased either of these subscription products, your existing benefits will remain fully intact and will not be affected in any way; please continue to use them with confidence. If you have any questions regarding this change, please feel free to contact customer service at any time.

Blended Token Explanation

The tokens within the subscription package are blended tokens, which can be used to call any model included in the package. The specific consumption ratio is based on real-time calculation during API calls.

• Benchmark Definition: The input token price of the claude-haiku-4-5-20251001 model serves as the benchmark price, i.e., $1/Mtoken (one million input tokens valued at $1 USD).
• Conversion Rule: When calling other models, the system will automatically convert and deduct blended tokens from the subscription package based on the ratio of that model's catalog price to the benchmark model's price. For instance, if a model's input price is $3/Mtoken, the number of blended tokens consumed when calling that model will be 3 times that of the benchmark model.
• Consumption Calculation: The token consumption for a single API call is Input token consumption + Cached writes token consumption + Cached reads token consumption + Output token consumption, all converted according to each model's deduction coefficient.

Example: In a single claude-sonnet-4-6 inference request

• Token Consumption: input tokens: 11, Cached writes (5m) tokens: 4629, output tokens: 2456
• Deduction Coefficients: input: 3, Cached writes (5m): 6.25, output: 25
• Then Blended tokens consumed = 11 × 3 + 4629 × 6.25 + 2456 × 25 ≈ 90,364