tencent cloud

LLM Service TokenHub

DokumentasiLLM Service TokenHubToken PlanEnterprise Token PlanEnterprise Pro Plan

Enterprise Pro Plan

Unduh
Mode fokus
Ukuran font
Terakhir diperbarui: 2026-09-15 14:56:06
Diterjemahkan oleh AI
The Token Plan Enterprise Pro Plan is a monthly prepaid subscription for large language model APIs designed for enterprises and teams. It employs a monthly prepaid model, supports custom monthly budgets and multi-key quota allocation, and centrally manages multi-model call quotas through a credit system, enabling more flexible budget control and team usage allocation.

Quick Start

Users who are already familiar with the Token Plan Enterprise Pro Plan can quickly get started by following the Quick Start guide.

Package Details

Package Specifications and Pricing

Level
Description
Package specification
Supports custom monthly purchase of credit quotas. A single purchase requires a minimum of 50,000 credits. Final prices and rules are subject to the console.
List price
The price for 50,000 credits is USD 70/month.

Purchase Notes

Category
Description
Purchase Instructions
The Token Plan Enterprise package takes effect immediately upon purchase and activation. Please create an API Key and start using it as soon as possible.
Quota Instructions
Credits not used during the applicable monthly billing period are forfeited at the end of that period and may not be carried forward, transferred, or redeemed in any subsequent period.
Renewal Instructions
Renew the package before it expires. Otherwise, the API Key will become invalid, and tools/applications/services using that API Key will immediately be unable to call the large model service. For details, see the Renewal Guide.

Available Models

The models supported in different regions are as follows:
Attention:
To continuously improve model service capabilities and user experience, the AI models included in the platform plans are provided as part of a dynamically updated model library. Models may be added, replaced, upgraded, adjusted in availability, or gradually discontinued based on factors such as model performance, service stability, compliance requirements, licensing status, and third-party model availability. The plans provide access to the corresponding models available in the platforms current model library, and do not constitute a commitment to the continuous, fixed, or permanent availability of any specific model.The models displayed at the time of subscription reflect availability at that point in time only. The actual available models, versions, and invocation scope are subject to the purchase page, console display, and platform announcements. For model discontinuations or major adjustments that may affect subscribed users, the platform will provide advance notice through reasonable means, such as announcements, in-site messages, or console notifications.
For the directly supplied model services provided directly by DeepSeek, TokenHub does not provide SLA guarantees. By using these models, you acknowledge and agree to comply with DeepSeek's service agreement. Please read the relevant terms carefully before use. If you do not accept the above terms, stop using the services immediately.
Singapore
Guangzhou
Model Name
Model ID
Remarks
Auto model
auto
-
GLM-5.3-Flash
glm-5.3-flash
-
GLM-5.3
glm-5.3
glm-5-3
-
GLM-5.2
glm-5.2
glm-5-2
-
MiniMax-M3
minimax-m3
minimax-m-3-0
-
Kimi K3
kimi-k3
-
Kimi K2.7 Code
kimi-k2.7-code
-
Kimi K2.7 Code HighSpeed
kimi-k2.7-code-highspeed
-
DeepSeek-V4-Flash
deepseek-v4-flash
-
DeepSeek-V4-Pro
deepseek-v4-pro
-
DeepSeek-V4-Flash 0731 GA
deepseek-v4-flash-0731
-
DeepSeek-V4-Pro 0813 GA
deepseek-v4-pro-0813
-
DeepSeek-V4.1-Flash (Vendor Direct)
deepseek/deepseek-flash
-
DeepSeek-V4-Flash 0731 GA (Vendor Direct)
deepseek-v4-flash-202605
deepseek/deepseek-v4-flash-0731
deepseek/deepseek-v4-flash
-
DeepSeek-V4-Pro 0813 Official Release (Vendor Direct)
deepseek-v4-pro-202606
deepseek/deepseek-v4-pro-0813
deepseek/deepseek-v4-pro
-
DeepSeek-V4-Flash-Vision-Exp (Vendor Direct)
deepseek/deepseek-v4-flash-vision-exp
-
Model Name
Model ID
Remarks
Auto model
auto
-
GLM-5.3-Flash
glm-5.3-flash
-
GLM-5.3
glm-5.3
glm-5-3
-
GLM-5.2
glm-5.2
glm-5-2
-
MiniMax-M3
minimax-m3
minimax-m-3-0
-
Kimi K3
kimi-k3
-
Kimi K2.7 Code
kimi-k2.7-code
-
Kimi K2.7 Code HighSpeed
kimi-k2.7-code-highspeed
-
DeepSeek-V4-Flash
deepseek-v4-flash
-
DeepSeek-V4-Pro
deepseek-v4-pro
-
DeepSeek-V4-Flash 0731 GA
deepseek-v4-flash-0731
-
DeepSeek-V4-Pro 0813 GA
deepseek-v4-pro-0813
-
DeepSeek-V4.1-Flash (Vendor Direct)
deepseek/deepseek-flash
-
DeepSeek-V4-Flash 0731 GA (Vendor Direct)
deepseek-v4-flash-202605
deepseek/deepseek-v4-flash-0731
deepseek/deepseek-v4-flash
-
DeepSeek-V4-Pro 0813 Official Release (Vendor Direct)
deepseek-v4-pro-202606
deepseek/deepseek-v4-pro-0813
deepseek/deepseek-v4-pro
-
DeepSeek-V4-Flash-Vision-Exp (Vendor Direct)
deepseek/deepseek-v4-flash-vision-exp
-

Access Address

For different regions, the platform provides different access addresses to ensure stable access for you. The default access addresses are as follows:
Attention:
Select the corresponding API address based on the region where the service is activated. Cross-region and cross-site service calls are not supported.
Region
Access Address
Resource Scheduling Scope
Guangzhou
Base URL
OpenAI API protocol: https://tokenhub.tencentcloudmaas.com/plan/v3
Anthropic API protocol: https://tokenhub.tencentcloudmaas.com/plan/anthropic
Complete URL
OpenAI API protocol: https://tokenhub.tencentcloudmaas.com/plan/v3/chat/completions
OpenAI Responses API protocol: https://tokenhub.tencentcloudmaas.com/plan/v3/responses
Anthropic Messages API protocol: https://tokenhub.tencentcloudmaas.com/plan/anthropic/v1/messages
Chinese mainland
Singapore
Base URL
OpenAI API protocol: https://tokenhub-intl.tencentcloudmaas.com/plan/v3
Anthropic API protocol: https://tokenhub-intl.tencentcloudmaas.com/plan/anthropic
Complete URL
OpenAI Chat Completions API protocol: https://tokenhub-intl.tencentcloudmaas.com/plan/v3/chat/completions
OpenAI Responses API protocol: https://tokenhub-intl.tencentcloudmaas.com/plan/v3/responses
Anthropic Messages API protocol: https://tokenhub-intl.tencentcloudmaas.com/plan/anthropic/v1/messages
Global

Credit Deduction Rules

Credits are the usage billing unit for the Token Plan Enterprise Pro Plan, used to quantify the consumption of model resources. All API Keys under the same plan share the plan's credit pool, and credits are deducted in real time based on actual call volume.

Deduction Formula

Credits consumed per request = (Number of cache-hit input tokens × Cache-hit input price + Number of cache-miss input tokens × Cache-miss input price + Number of output tokens × Output price) / 1,000,000.

Ctedit Deduction Price

Credit redemption prices vary by region, as detailed below. To help users intuitively assess "how many tokens a certain amount of credits can purchase", the platform calculates the comprehensive unit price and the number of redeemable tokens based on operational experience values from June 2026. You can refer to the "Estimated Value" column.
Note:
Notes on Estimated Value Calculation:
This calculation result serves only as a reference for enterprise budget planning and does not represent the actual number of tokens available for use. The actual credits consumed and the number of usable tokens will be affected by factors such as the cache hit rate, input/output token ratio, model mix usage, and real-time pricing rules in real business scenarios. The final determination is based on the actual call results. Please note the actual credits consumed.

Comprehensive Unit Price Calculation Formula
Comprehensive unit price = (Cache hit rate × Cache-hit input price + (1 - Cache hit rate) × Cache-miss input price) × Input proportion + Output price × Output proportion.
Among them:
Cache hit rate: Calculated based on historical operational data of each model and already incorporated into the comprehensive unit price calculation.
Input proportion: 20/21;
Output proportion: 1/21;
If the prices displayed on the page involve rounding, the calculation results may slightly differ from those derived using the formula.
Singapore
Guangzhou
Model
Tier Condition
Input Price (Cache Hit)
(Credits/Million tokens)
Input Price (Cache Miss)
(Credits/Million tokens)
Output Price
(Credits/Million tokens)
Estimated Price
Estimated Comprehensive Unit Price
(points/million tokens)
Estimated tokens deductible with 500k points
(billion tokens)
Estimated tokens deductible with 1 million points
(billion tokens)
Auto model
-
51
334
1650
Approximately 191
Approximately 26.18
Approximately 52.36
GLM-5.3-Flash
-
21.429
107.143
357.143
Approximately 46
Approximately 108.70
Approximately 217.39
GLM-5.3
-
185.714
1000
3142.857
Approximately 412
Approximately 12.14
Approximately 24.27
GLM-5.2
-
185.714
1000
3142.857
Approximately 412
Approximately 12.14
Approximately 24.27
MiniMax-M3
Input [0, 512k)
42.857
214.286
857.143
Approximately 95
Approximately 52.63
Approximately 105.26
Input 512k+
85.714
428.571
1714.286
Approximately 189
Approximately 26.46
Approximately 52.91
Kimi K3
-
214.286
2142.857
10714.286
Approximately 861
Approximately 5.81
Approximately 11.61
Kimi K2.7 Code
-
135.714
678.571
2857.143
Approximately 296
Approximately 16.89
Approximately 33.78
Kimi K2.7 Code HighSpeed
-
271.429
1357.143
5714.286
Approximately 861
Approximately 5.81
Approximately 11.61
DeepSeek-V4-Flash
-
20
100
200
Approximately 61
Approximately 81.97
Approximately 163.93
DeepSeek-V4-Pro
-
103.571
1242.857
2485.714
Approximately 532
Approximately 9.40
Approximately 18.80
Attention:
All DeepSeek V4 [Vendor Direct] models will adjust their peak/off-peak billing rules in sync with the vendor. Starting from 00:00 Beijing Time on August 29, 2026 (Saturday), the original peak/off-peak billing will continue to apply on weekdays (Monday to Friday), with peak hours from 9:00 to 12:00 and 14:00 to 18:00 Beijing Time and all other hours being off-peak. On weekends (Saturday and Sunday), peak and off-peak hours will no longer be distinguished, and billing will be based on the off-peak price for the entire day.
DeepSeek V4 GA Peak/Off-Peak Billing Rules: Peak hours are Monday to Sunday 9:00–12:00 and 14:00–18:00 Beijing Time (all other hours are off-peak).
Peak and Off-Peak Period Determination Rules: The billing period for a single request is determined by the time (Beijing Time) when the platform server receives that request. If the request reception time falls within peak hours, that request is billed at the peak rate. If it falls within off-peak hours, it is billed at the off-peak rate. The request processing duration and return time do not affect the period determination.

Model
Condition
Input Price (Cache Hit)
(Credits/Million tokens)
Input Price (Cache Miss)
(Credits/Million tokens)
Output Price
(Credits/Million tokens)
Estimated Price
Estimated Comprehensive Unit Price
(points/million tokens)
Estimated tokens deductible with 500k points
(billion tokens)
Estimated tokens deductible with 1 million points
(billion tokens)
DeepSeek-V4-Flash 0731 GA
OFF-PEAK
5
157.143
471.429
Approximately 55
Approximately 90.91
Approximately 181.82
PEAK
10
314.286
942.857
Approximately 109
Approximately 45.87
Approximately 91.74
DeepSeek-V4-Pro 0813 GA
OFF-PEAK
15.714
471.429
1414.286
Approximately 134
Approximately 37.31
Approximately 74.63
PEAK
31.429
942.857
2828.571
Approximately 269
Approximately 18.59
Approximately 37.17
DeepSeek-V4.1-Flash (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82
DeepSeek-V4-Flash 0731 GA (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82
DeepSeek-V4-Pro 0813 GA
(Vendor Direct)
OFF-PEAK
15.714
471.429
1414.286
Approximately 134
Approximately 37.31
Approximately 74.63
PEAK
31.429
942.857
2828.571
Approximately 269
Approximately 18.59
Approximately 37.17
DeepSeek-V4-Flash-Vision-Exp (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82
Model
Tier Condition
Input Price (Cache Hit)
(Credits/Million tokens)
Input Price (Cache Miss)
(Credits/Million tokens)
Output Price
(Credits/Million tokens)
Estimated Price
Estimated Comprehensive Unit Price
(points/million tokens)
Estimated tokens deductible with 500k points
(billion tokens)
Estimated tokens deductible with 1 million points
(billion tokens)
Auto model
-
51
334
1650
Approximately 191
Approximately 26.18
Approximately 52.36
GLM-5.3-Flash
-
22.829
79.4
277.907
Approximately 40
Approximately 125.00
Approximately 250.00
GLM-5.3
-
198.5
794
2779.071
Approximately 384
Approximately 13.02
Approximately 26.04
GLM-5.2
-
200
800
2800
Approximately 387
Approximately 12.92
Approximately 25.84
MiniMax-M3
Input [0, 512k)
42.857
214.286
857.143
Approximately 95
Approximately 52.63
Approximately 105.26
Input 512k+
85.714
428.571
1714.286
Approximately 189
Approximately 26.46
Approximately 52.91
Kimi K3
-
195.071
1950.714
9752.143
Approximately 784
Approximately 6.38
Approximately 12.76
Kimi K2.7 Code
-
135.714
678.571
2857.143
Approximately 296
Approximately 16.89
Approximately 33.78
Kimi K2.7 Code HighSpeed
-
271.429
1357.143
5714.286
Approximately 861
Approximately 5.81
Approximately 11.61
DeepSeek-V4-Flash
-
20
100
200
Approximately 61
Approximately 81.97
Approximately 163.93
DeepSeek-V4-Pro
-
103.571
1242.857
2485.714
Approximately 532
Approximately 9.40
Approximately 18.80
Attention:
All DeepSeek V4 [Vendor Direct] models will adjust their peak/off-peak billing rules in sync with the vendor. Starting from 00:00 Beijing Time on August 29, 2026 (Saturday), the original peak/off-peak billing will continue to apply on weekdays (Monday to Friday), with peak hours from 9:00 to 12:00 and 14:00 to 18:00 Beijing Time and all other hours being off-peak. On weekends (Saturday and Sunday), peak and off-peak hours will no longer be distinguished, and billing will be based on the off-peak price for the entire day.
DeepSeek V4 GA Peak/Off-Peak Billing Rules: Peak hours are Monday to Sunday 9:00–12:00 and 14:00–18:00 Beijing Time (all other hours are off-peak).
Peak and Off-Peak Period Determination Rules: The billing period for a single request is determined by the time (Beijing Time) when the platform server receives that request. If the request reception time falls within peak hours, that request is billed at the peak rate. If it falls within off-peak hours, it is billed at the off-peak rate. The request processing duration and return time do not affect the period determination.

Model
Condition
Input Price (Cache Hit)
(Credits/Million tokens)
Input Price (Cache Miss)
(Credits/Million tokens)
Output Price
(Credits/Million tokens)
Estimated Price
Estimated Comprehensive Unit Price
(points/million tokens)
Estimated tokens deductible with 500k points
(billion tokens)
Estimated tokens deductible with 1 million points
(billion tokens)
DeepSeek-V4-Flash 0731 GA
OFF-PEAK
5
157.143
471.429
Approximately 55
Approximately 90.91
Approximately 181.82
PEAK
10
314.286
942.857
Approximately 109
Approximately 45.87
Approximately 91.74
DeepSeek-V4-Pro 0813 GA
OFF-PEAK
15.714
471.429
1414.286
Approximately 134
Approximately 37.31
Approximately 74.63
PEAK
31.429
942.857
2828.571
Approximately 269
Approximately 18.59
Approximately 37.17
DeepSeek-V4.1-Flash (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82
DeepSeek-V4-Flash 0731 GA (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82
DeepSeek-V4-Pro 0813 GA
(Vendor Direct)
OFF-PEAK
15.714
471.429
1414.286
Approximately 134
Approximately 37.31
Approximately 74.63
PEAK
31.429
942.857
2828.571
Approximately 269
Approximately 18.59
Approximately 37.17
DeepSeek-V4-Flash-Vision-Exp (Vendor Direct)
OFF-PEAK
2.143
107.143
428.571
Approximately 27
Approximately 185.19
Approximately 370.37
PEAK
4.286
214.286
857.143
Approximately 55
Approximately 90.91
Approximately 181.82

Quotas and Limits

Quota

Quota
Description
API Key creation quota
Multiple API Keys can be created under each package, and one API Key can be created per 10,000 credits for each package.
API Key configuration modification quota
Each API Key can be modified a maximum of 10 times per day.

Limit

Downgrading is not supported for the Enterprise Pro Plan.
The Enterprise Pro Plan does not support cancellation once purchased.

More Operations

For more operations, see the Operation Guide.
Topik sebelumnya: Enterprise Token Plan OverviewTopik selanjutnya: Enterprise Lite Plan

Bantuan dan Dukungan

Apakah halaman ini membantu?

masukan

Search
    Enter keywords to search documentation