Skip to main content
Qwen Cloud home page
Docs
Token Plan

FAQ

Fix Token Plan issues fast

What is the difference between Token Plan and Coding Plan?

Token PlanCoding Plan
Use casesDay-to-day work for solo founders, teams, and enterprisesIndividual development scenarios
Supported modelsText generation and image generation modelsText generation models
Billing methodCredits deducted based on token consumptionBilled per model call
Usage frequencyUnlimitedQuota per 5 hours / per week
API Key and Base URLGet from the Token Plan page. See Quick startGet from the Coding Plan page
Peak-time performanceMulti-tenant isolationRequests may queue during peak hours
Data securityYour data is not used to train modelsUser data authorization required

Important resources

Start here: 60-second diagnosis

If your tool is not working and you see errors like 401, 403, 404, invalid access token, invalid api-key, quota exceeded, or connection failures, check these first:
  1. API key: Must be the Token Plan dedicated key, not a pay-as-you-go sk- key or a Coding Plan sk-sp- key
  2. Base URL: Must contain token-plan, not dashscope-intl or coding
  3. Authentication header: Must use Authorization: Bearer, not x-api-key
  4. Subscription status: Check whether your subscription is active and credits are not depleted
  5. Updated config: Restart the tool after updating the API key or base URL
If you are still blocked, continue with the questions below.

Connection and usage

Which text generation models are supported?

Token Plan supports the following text generation models:
  • qwen3.7-plus
  • qwen3.7-max
  • qwen3.6-plus
  • qwen3.6-flash
  • deepseek-v4-pro
  • deepseek-v4-flash
  • deepseek-v3.2
  • kimi-k2.6
  • kimi-k2.5
  • glm-5.1
  • glm-5
  • MiniMax-M2.5
For the complete and up-to-date list, see Supported models.

Common errors and solutions

Error messagePossible causesSolution
401 InvalidApiKey: No API-key provided.The request did not include the API key in either the Authorization: Bearer or x-api-key header.Go to the Token Plan page, copy your dedicated API key, and configure it in your tool.
401 InvalidApiKey: Invalid API-key provided.1. You used a Qwen Cloud pay-as-you-go key (sk-xxx) or a Coding Plan key (sk-sp-xxx). 2. Your Token Plan subscription has expired. 3. The API key is incomplete or contains spaces.1. Confirm you are using the Token Plan dedicated API key. Ensure the key is copied completely without any spaces. 2. Verify whether the subscription has expired. 3. If the error persists, reset the API key, then configure the new key.
400 Model not exist.The model name does not match any available model.Check the model name against Supported models and ensure it is spelled exactly as documented.
model 'xxx' not found or not supportedThe model name is misspelled or uses the wrong case.Ensure the model name is case-sensitive and matches the model ID listed in Supported models.
400 InvalidParameter: url error, please check url!The Base URL path does not match the protocol. For example, the OpenAI-compatible path is configured on the Anthropic endpoint, or vice versa.Choose the endpoint matching the protocol your tool uses: Anthropic-compatible (Claude Code and similar) ends with /apps/anthropic. OpenAI-compatible (Cursor, Qwen Code, and similar) ends with /compatible-mode/v1.
400 InvalidParameter: Range of max_tokens should be [1, xxxx]The max_tokens value in the request (or the maximum output length in the tool configuration) exceeds the maximum output tokens supported by the current model.Set max_tokens to a value not exceeding the limit shown in the error message.
400 invalid_parameter_error: The thinking_budget parameter must be a positive integer and not greater than xxxxxThe thinking length configured in the tool (such as thinking_budget or budgetTokens) exceeds the limit supported by the current model.Set the thinking length to a value not exceeding the limit shown in the error, or remove the parameter for models that do not support thinking mode.
invalid access token or token expiredYou used the base URL for the Coding Plan or another plan.Use the correct Token Plan base URL. OpenAI compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1. Anthropic compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/apps/anthropic.
Incorrect API key providedYou used the general Qwen Cloud base URL (dashscope-intl.aliyuncs.com).Use the correct Token Plan base URL. OpenAI compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1. Anthropic compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/apps/anthropic.
Range of input length should be [1, xxx]Your input (including conversation history and code context) exceeds the model's maximum context length.Start a new session or shorten the context. You can also switch to a model with a larger context window.
429 API-Key Requests rate limit exceededRequests were too dense within a short period and triggered the model rate limit.Wait one minute and retry. If this occurs frequently, reduce the request frequency and verify the API key is not being shared with others.
429 Throttling.AllocationQuota / insufficient_quotaThe plan's tokens are depleted -- both the seat quota and the shared quota pack are exhausted.Wait for the quota to reset at the next billing cycle, or purchase a shared usage package to add more credits.
Connection errorThe base URL domain is misspelled, or you have a network connection issue.Check the spelling of the base URL domain and your network connection.

Image generation

Which image generation models are supported?

Token Plan supports the following image generation models:
  • qwen-image-2.0
  • qwen-image-2.0-pro
  • wan2.7-image
  • wan2.7-image-pro
For details, see Supported models.

Can I use image models in coding tools?

Image generation models use separate APIs and cannot be called directly through the text model's base URL. You must integrate them through your tool's skill or extension mechanism. For configuration steps, see the image generation model integration section in each tool's setup guide.

How do I use image generation models in coding tools?

Most coding tools (such as Claude Code, Cursor, and Cline) support extension mechanisms like Skills, Slash Commands, or MCP servers. You can use these mechanisms to call image generation models within your coding workflow. For step-by-step instructions, see Integrate multimodal generation models.

Common errors & fixes

Which API key and base URL should I use?

Use a Token Plan dedicated key and a Token Plan base URL:
  • API key: Get it from the Token Plan page
  • OpenAI compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1
  • Anthropic compatible: https://token-plan.ap-southeast-1.maas.aliyuncs.com/apps/anthropic
Pay-as-you-go dashscope-intl URLs and Coding Plan coding-intl URLs do not work with Token Plan keys.

I got 401 or 403. Is my API key invalid?

If you see 401 invalid access token, InvalidApiKey, or 403 invalid api-key, the problem is usually a wrong key type, an expired subscription, a malformed key, or a key and URL mismatch. Check these points:
  1. Confirm the key is your Token Plan dedicated key
  2. Re-copy the key to remove hidden spaces or line breaks
  3. Confirm your subscription is active on the Token Plan page
  4. Make sure the base URL contains token-plan
In most cases:
  • 401 invalid access token means wrong key type, expired subscription, or malformed key
  • 403 invalid api-key means you are using a Token Plan key with a non-Token Plan URL
  • Incorrect API key provided means you are using a Token Plan key with the general dashscope-intl URL

Why do I get 404 or connection errors?

If you get 404 status code (no body), endpoint not found, or connection errors, the base URL path is usually wrong for your tool. Use the correct path for your protocol:
  • Anthropic-compatible tools use /apps/anthropic
  • OpenAI-compatible tools use /compatible-mode/v1
Also check that the domain is exactly token-plan.ap-southeast-1.maas.aliyuncs.com.

Why is my model "not supported"?

If you get model 'xxx' is not supported or model 'xxx' not found, the model ID is usually incorrect, case-sensitive, or copied with extra spaces. Check these points:
  1. Use the exact model name from Supported models
  2. Remove any leading or trailing spaces
  3. Verify capitalization exactly as documented

How to handle input length limits?

If you get 400 InvalidParameter: Range of input length should be [1, xxx], your context or request exceeds model limits. Try these fixes:
  1. Create a new session
  2. Switch to a model with a larger context window
  3. In your tool, configure context length limits if available

What should I do if I see data_inspection_failed?

If you see data_inspection_failed, refer to Troubleshooting data inspection errors.

Product features

Can I mix Token Plan, Coding Plan, and pay-as-you-go keys?

No. The API keys and base URLs for Token Plan, Coding Plan, and Qwen Cloud pay-as-you-go are not interchangeable. Do not mix them. Using an incorrect API key will not deduct credits from your Token Plan quota.

Can I use one subscription across multiple tools?

Yes. You can use the same API key in all compatible AI tools, and all usage draws from the same quota.

Are there usage restrictions?

The service is for interactive use in compatible AI programming and agent tools only. It cannot be used for automated scripts or application backends. Violating these terms may result in subscription suspension or revocation of your API key. All supported models in Token Plan are full-featured, unquantized versions.

How do I reset or manage my API key?

Use the Reset button on the API Keys page to reset your key. Keep these rules in mind:
  • After reset, update the key in all tools
  • If you resubscribe after expiration, the key changes
  • Token Plan supports one key per subscription

Billing and quota

Why am I still billed after subscribing?

If you still see pay-as-you-go charges after subscribing to Token Plan, it is usually one of the following:
  • Wrong credentials (most common): Your tool is using a pay-as-you-go API key (sk- format) and a base URL without token-plan. Use your Token Plan dedicated key and ensure the base URL contains token-plan.
  • Tool falling back to general credentials: If your tool has both general and Token Plan credentials configured, it may route requests through the general credentials. Remove the general API configuration and keep only Token Plan credentials.
  • Client cache not cleared: After updating credentials, clear the tool's cache and restart it to ensure it uses the new configuration.
Use the Token Plan page to verify usage. If needed, reset your key on the API Keys page.

How are credits deducted?

Credit consumption is calculated based on the input, cached, and output tokens in each request. The system first deducts credits from your seat quota. After the seat quota is depleted, credits are deducted from the shared usage package. If all credits are used, the service is suspended until the next billing cycle or until you purchase a shared usage package to add more credits.

What happens when my quota runs out?

Once your seat quota is depleted, the system automatically begins deducting from the shared usage package. If you run out of credits entirely, the service is suspended. You can resume the service in one of the following ways:
  • Purchase a shared usage package to add more credits
  • Wait for your quota to automatically reset at the start of the next billing cycle
There is no automatic fallback to pay-as-you-go when credits are exhausted.

When do my credits reset?

The seat quota resets at the start of each new subscription month. Renewing or resubscribing mid-cycle does not immediately add or reset credits — the renewed quota becomes available only when the next billing cycle begins. Unused credits do not roll over to the next month. Similarly, unused credits in the shared usage package expire at the end of each billing cycle.

I just renewed my plan but my credits are still 0. Why?

Credits reset only at the start of each subscription month, not at the moment of renewal. If you renew mid-cycle, your renewal extends the subscription, but the new quota does not take effect until the next billing cycle begins. If you need credits before the cycle resets, purchase a shared usage package.

How can I view my usage?

Check the Token Plan page for overall usage, including seat quota and shared usage package details.

Can I upgrade or downgrade my plan?

No. Switching between seat tiers is not supported. If you need more credits, purchase a shared usage package.

Can I buy multiple subscriptions?

No. Each Qwen Cloud account is limited to one Token Plan subscription.

Can I get a refund?

No. Token Plan does not support refunds. Please review the Usage rules for details before you subscribe.

Does an overdue payment affect my plan?

Token Plan is a prepaid subscription product. As long as your plan's credits are not depleted and the subscription is still valid, an overdue payment on your Qwen Cloud account will not affect your use of the service.

Can multiple people share one key?

No. Token Plan API keys are for the subscriber's personal use only. Public exposure or account sharing may result in automatic key suspension.

Data security

How is my data protected?

Token Plan does not use your conversation data for model training.