Answers to common questions about Token Plan editions, purchases and subscriptions, cancellations and refunds, usage limits, and integration.
Product and editions
What is the difference between Personal Edition and Team Edition?
Personal Edition is for individual developers and is subscribed by plan tier. Team Edition is for teams and enterprises and is subscribed by seat, with seat management, member usage analysis, and SSO integration. The two editions also differ in quota mechanism, data security, and peak-hour performance. For a full comparison, see Personal Edition subscription and benefits and Team Edition subscription and benefits.
Can I purchase both Personal Edition and Team Edition? Are quotas calculated separately or shared?
Yes. The same Alibaba Cloud account can hold both Personal Edition and Team Edition at the same time. Each is billed independently, and quotas are not shared.
Can I upgrade or convert Personal Edition to a Team Edition seat?
No. Personal Edition and Team Edition are separate products with no shared quota, and Personal Edition cannot be upgraded or converted directly to a Team Edition seat. To use Team Edition, subscribe to it separately. Personal Edition does not currently support unsubscribing; you can stop renewing when the subscription expires.
Can a Team Edition API key be used with Personal Edition?
No. Personal Edition and Team Edition each generate independent API keys that cannot be used interchangeably. The system automatically identifies the corresponding plan based on the API key.
If I already have Personal Edition, will purchasing Team Edition conflict?
No. Both can be held at the same time and are billed independently. The corresponding plan is automatically matched based on the API key.
Which models does Token Plan support?
Token Plan supports Qwen, Wan, DeepSeek, Moonshot AI, Zhipu AI, MiniMax, and HappyHorse models. The model range differs between Personal Edition and Team Edition. For the full list, see the models supported by Personal Edition and Team Edition.
Does Token Plan support third-party tools such as Cursor, Claude Code, and Cline?
Yes. Token Plan is compatible with the OpenAI and Anthropic protocols. Any tool that supports a custom Base URL and API key can connect. For configuration, see Quick start.
What are built-in model tools?
Built-in model tools are capabilities provided by the model, including web search, text-to-image search, reverse image search, web scraping, and code interpreter. They are automatically triggered when you call a model through the Responses API and are deducted from Credits. Both Personal Edition and Team Edition support them. For details, see Built-in model tools.
Does Token Plan provide a free trial quota or free Tokens?
The Token Plan itself does not provide a trial quota or free Tokens. Model Studio provides a separate free quota for some models; you can check the console for the list of models with available free quota.
Purchase and subscription
Can RAM users use Token Plan?
Yes. RAM users need Token Plan permissions and Billing Management permissions granted by the primary account: the Billing Management policy is AliyunBSSReadOnlyAccess for Personal Edition and AliyunBSSFullAccess for Team Edition. For complete authorization steps and a permission comparison, see Permissions and usage policy · RAM user authorization.
RAM authorization and seat assignment are two independent concepts: RAM authorization determines who can manage the subscription; seat assignment determines who can use the models. Team Edition members still need to be assigned a seat in team management before they can obtain an API key.
Can I upgrade? How is quota calculated after an upgrade?
Upgrading from a lower tier to a higher tier is supported. Upgrades do not reset the current subscription cycle: you pay the prorated price difference for the remaining days, and additional Credits for the current cycle are prorated based on the monthly quota difference between the new and old tiers. The new tier's monthly quota applies from the next subscription month.
For the formula and an example, see Usage and quota · Upgrade quota calculation.
Can I downgrade?
Downgrading is not supported. To switch to a lower tier or seat type, repurchase or resubscribe after the subscription expires. For details, see Personal Edition · Other and Team Edition · Important notes.
How do I cancel auto-renewal?
- Personal Edition: Sign in to the Model Studio console Token Plan page and disable auto-renewal in subscription management.
- Team Edition: Sign in to Billing Management > Renewal Management, find the Token Plan Team Edition order, and disable auto-renewal.
Do I need to reconfigure the API key after renewal?
No. Renewal, repurchasing after the subscription expires, and repurchasing after unsubscribing from a seat do not change the API key of an assigned seat. The API key is temporarily unavailable while the subscription is expired or unsubscribed, and becomes usable again after renewal or repurchase succeeds.
How do I view the effective time and remaining days of a plan?
- Effective time: Sign in to the Model Studio console and view the Token Plan page.
- Remaining days: The remaining days shown in the console are rounded down to complete 24-hour periods. For example, 22 days and 20 hours remaining is displayed as 22 days; the actual validity is based on the end time.
Does Token Plan support student vouchers?
No. Student vouchers apply only to the products specified on the promotion page and cannot be used to purchase Token Plan.
Cancellation and refunds
Can I unsubscribe from Personal Edition?
Personal Edition does not currently support unsubscribing. You can stop renewing when the subscription expires.
Can I unsubscribe from Team Edition seats?
Yes, you can unsubscribe by seat. Seats with existing usage consumption cannot be unsubscribed. Refunds are returned to the original payment account within an estimated 1-3 business days. For the operation, see Team Edition · Unsubscribe seats.
Usage limits
What does "production API automation is prohibited" mean?
Token Plan Personal Edition is for individual interactive development through officially designated tools (such as Cursor, Claude Code, and Windsurf). Using the API key for production automated services, batch scripts, or scheduled background tasks is not allowed.
Can team members view team usage?
No. Team members cannot view the team subscription, seat quota, or shared usage pack balance; the owner or administrator must view them. For details, see Team management · Roles and permissions.
Is data used to train models?
Personal Edition data usage is governed by the service agreement; Team Edition does not use conversation data for model training. For details, see Permissions and usage policy · Data usage policy.
Can multiple people share one account?
No. Token Plan Personal Edition is limited to a single user; sharing the same account or API key among multiple people is not allowed. For multi-user collaboration, use Token Plan Team Edition.
Can I use the same API key on multiple devices?
Yes. The same API key can be configured on multiple devices you own, but only for your own use. Resetting the API key immediately invalidates the old key, and you must then update all devices with the new key. Reset the key only if it is leaked.
How many concurrent Agents are supported?
Personal Edition concurrency depends on the plan tier:
| Tier | Recommended concurrency |
|---|---|
| Lite | 1-2 Agents |
| Essential | 2-3 Agents |
| Standard | 3-4 Agents |
| Pro | 6-8 Agents |
Will responses slow down during peak hours?
Personal Edition may queue during peak hours. For more stable throughput, upgrade to a higher tier or use Team Edition. Team Edition uses a multi-tenant isolation architecture and does not queue during peak hours.
What is the rate limit (TPM/RPM)? Can it be increased?
Rate limits are adjusted dynamically based on overall load to keep the service stable. Plan rate limits cannot be increased. If you are rate-limited, trim the context, reduce task complexity, and wait about 1 minute before retrying.
Does Team Edition have concurrency limits?
Yes. The platform dynamically adjusts the concurrency limit based on overall resource load. If you reach the concurrency limit, wait a moment and retry.
Integration and errors
Distinguish four scenarios first
| Symptom | Cause | Handling |
|---|---|---|
| 429 insufficient_quota | Plan quota or usage pack quota exhausted | Wait for reset, upgrade, or purchase a usage pack or shared usage pack |
| 429 Requests rate limit exceeded | Requests are too concentrated in a short period | Wait about 1 minute before retrying and reduce the request frequency |
| Concurrency limit | The number of simultaneously running Agents exceeds the recommended value for the tier | Reduce the number of simultaneously running Agents, or upgrade to a higher tier |
| 401, 403, 404 | API key, Base URL, or model configuration error | Check the key prefix, Base URL, and model ID |
Troubleshooting order
- Confirm that the API key starts with
sk-sp-and is used with the Token Plan Base URL. - Confirm that the model ID is in the models supported by Personal Edition and Team Edition and applies to the current edition.
- Check whether the quota is exhausted on the My Subscriptions and Usage Analysis pages in the console.
- Check whether the error is
insufficient_quota,Requests rate limit exceeded, or an authentication error.
Common errors
| Error message | Possible cause | Solution |
|---|---|---|
401 InvalidApiKey: No API-key provided. | No API key is included in the request header | Generate an API key and configure it in the tool |
401 InvalidApiKey: Invalid API-key provided. | Used a pay-as-you-go API key or a Coding Plan key by mistake; subscription expired; key copied incompletely | Confirm that you are using the Token Plan API key, and ensure that it is complete and contains no spaces |
404 model_not_found: Model not exist. | Model name is misspelled or not in the supported list | Confirm that the model name is case-sensitive and matches a model ID supported by the plan |
404 model 'xxx' not found or not supported | Model name is misspelled or not in the supported list | Confirm that the model name is case-sensitive and matches a model ID supported by the plan |
403 AccessDenied.Unpurchased: Access to model denied. | The model is supported only by Team Edition, not Personal Edition | Switch to a model supported by Personal Edition, or use Token Plan Team Edition |
401 invalid access token or token expired | Used the Coding Plan or another billing mode Base URL by mistake | Use the Token Plan Base URL |
401 Incorrect API key provided | Used the general Model Studio Base URL (dashscope-intl.aliyuncs.com) by mistake | Use the Token Plan Base URL |
400 Range of input length should be [1, xxx] | Input content exceeds the model's maximum context length | Start a new session to clear history, or use the tool's context compression command |
400 invalid_parameter_error: Unexpected item type in content. | Non-text content such as an image was passed to a text-only model | Switch to a model that supports the modality, or confirm that the request content matches the model capabilities |
400 error when calling image or video generation models | Image and video generation models use separate endpoints and cannot be called directly through the text model Base URL | Integrate through the tool's Skill, Slash Command, or Agent extension mechanism |
429 Requests rate limit exceeded | Requests are too concentrated in a short period; multiple parallel Agents may also trigger the concurrency limit | Wait about 1 minute before retrying and reduce the request frequency and concurrency |
429 insufficient_quota: Your token-plan quota has been exhausted. | Plan quota or usage pack quota exhausted | Purchase a usage pack, upgrade the plan, or wait for the next subscription month to reset the quota |
502 Bad Gateway | Intermittent server-side error | Retry later |
Quota exhausted
A 429 insufficient_quota response indicates that the quota is exhausted. For the deduction order and recovery options, see Usage and quota · Quota exhaustion and recovery.
Rate limiting and concurrency
A 429 Requests rate limit exceeded response indicates rate limiting, not quota exhaustion. Rate limits are adjusted dynamically based on overall load, and plan rate limits cannot be increased. Wait about 1 minute before retrying, trim the context, reduce concurrency, or lower the request frequency. Team Edition also has concurrency limits.
Authentication and configuration
Token Plan, Coding Plan, and pay-as-you-go API keys and Base URLs are isolated and must be used in matching pairs:
- Token Plan: API key starting with
sk-sp-+ Token Plan Base URL. - Coding Plan: Coding Plan API key +
https://coding-intl.dashscope.aliyuncs.com/v1. - Pay-as-you-go: Model Studio API key (starting with
sk-) +https://dashscope-intl.aliyuncs.com/compatible-mode/v1.