Token Plan Personal Edition is an AI large model subscription service for individual developers. It uses Credits as a unified billing unit and supports text models, multimodal models, and Harness tools. It is compatible with mainstream AI coding and agent tools.
Key features
- Unified Credits billing: Use Credits to pay for different models and Harness tools under a single billing unit.
- Multimodal model support: Covers text generation, reasoning, visual understanding, image generation, speech synthesis, real-time voice conversation, speech recognition, video generation, and more.
- Harness tool integration: Supports web search, text-to-image search, image-to-image search, web scraping, and code interpreter.
- Usage packs: Purchase usage packs after exceeding your quota to continue using the service without restrictions.
- Compatible with multiple tools: Works with mainstream AI coding and agent tools such as Claude Code, Cursor, Qwen Code, Qoder, Qoder CN, and OpenClaw.
Plans and pricing
Lite plan | Essential plan | Standard plan | Pro plan | |
|---|---|---|---|---|
Pricing | Original price $8/month | Original price $16/month | Original price $25/month | Original price $80/month |
Monthly quota | 11,500 Credits | 25,500 Credits | 45,000 Credits | 180,000 Credits |
Concurrent Agents | 1-2 | 2-3 | 3-4 | 6-8 |
Benefits | Multimodal models Harness tools Mainstream tools | All Lite benefits | All Lite benefits Harness benefits | All Standard benefits Higher concurrency Harness benefits |
- Monthly quota: A 30-day subscription cycle starts from the subscription date. When cumulative consumption within the subscription cycle reaches the plan quota, the service is paused until the next subscription month. After the monthly quota is used up, purchase usage bundles as a supplement, or wait for the next subscription month to reset.
- Extra Bundle: Priced at $15/bundle/month, with 20,000 Credits per bundle. Requires an active subscription before purchase, hold up to 5 at the same time; after the plan's monthly quota is used up, usage bundle Credits are deducted automatically.
Credits billing mechanism
- High consumption per call: Credits for video generation increase with the duration and resolution, which may quickly use up a large part of your quota. For your first attempts, use a shorter duration and a lower resolution, check the actual consumption per call on the subscription page usage details in the console, and then decide on further usage.
- Batch settlement for asynchronous tasks: Credits for asynchronous tasks such as video generation are settled after the task completes, rather than deducted at submission. If you submit multiple asynchronous tasks within a short period, the Credits settled together after all the tasks complete may cause you to reach the quota limit quickly.
Billing description
The Credits consumed per request are dynamically determined by model type, token usage, thinking mode, and tool calls. Refer to the subscription page usage details in the console for actual consumption.
Deduction order
- Credits consumed by each call are deducted from the plan quota of the current subscription month first.
- After the plan's monthly quota is used up, usage bundle Credits are automatically deducted. When usage bundle Credits are also used up or not held, the service is paused. You can upgrade to a higher-tier plan, or wait for the quota to reset at the start of the next subscription month.
Quota reset
The quota resets automatically per subscription month: at the start of each subscription month, the plan quota is restored to the full monthly quota of the corresponding tier. When the weekly quota was removed on September 22, 2026, the remaining quota of existing subscriptions was reset once to the full monthly quota of their plan tier on that day, with subscription periods and expiration dates unchanged.
Supported models
Brand | Model ID | Capability |
|---|---|---|
Qwen | auto | Reasoning, text generation |
qwen3.8-max | Reasoning, visual understanding, text generation | |
qwen3.8-flash | Reasoning, visual understanding, text generation | |
qwen3.7-max | Reasoning, text generation | |
qwen3.7-plus | Reasoning, visual understanding, text generation | |
qwen3.6-flash | Reasoning, visual understanding, text generation | |
qwen-image-3.0-pro | Image generation | |
qwen-audio-3.0-tts-plus | Speech synthesis | |
qwen-audio-3.0-realtime-plus | Real-time voice conversation | |
qwen-audio-3.0-asr-flash | Speech recognition | |
Wanx | wan2.7-image | Image generation |
wan2.7-image-pro | Image generation | |
wan3.0-video | Video generation | |
DeepSeek | deepseek-v4.1-flash | Reasoning, visual understanding, text generation |
deepseek-v4-pro | Reasoning, text generation | |
deepseek-v4-pro-0813 | Reasoning, text generation | |
deepseek-v4-flash-0731 | Reasoning, text generation | |
Zhipu AI | glm-5.3 | Reasoning, text generation |
glm-5.2 | Reasoning, text generation | |
HappyHorse | happyhorse-1.1-i2v | Video generation |
happyhorse-1.1-t2v | Video generation | |
happyhorse-1.1-r2v | Video generation |
Supported Harness tools
Capability | Tool name |
|---|---|
Web search | web_search |
Text-to-image search | web_search_image |
Image-to-image search | image_search |
Web scraping | web_extractor |
Code interpreter | code_interpreter |
Subscription management
Upgrade
You can upgrade from a lower-tier plan to a higher-tier plan. Upgrades do not reset the current subscription cycle: you pay the prorated price difference for the remaining days of the current cycle, the current cycle only receives additional Credits prorated based on the quota difference, and the new tier's monthly quota applies from the next subscription month.
The Credits granted upon upgrade are not the full monthly quota of the new plan tier. Instead, they are prorated based on the price difference you actually paid and the number of days remaining in your current billing cycle:
Renewal
You can switch to a different renewal duration and renew for multiple cycles at once.
- Manual renewal: You can choose a different renewal duration and renew for multiple cycles.
- Auto-renewal: When enabled, the system automatically deducts fees and renews before expiration.
Other
- The Personal Edition does not currently support unsubscription.
- If you repurchase after your subscription expires, your API Key will change and you will need to reconfigure it in your tools. You can find your new API Key in the My Subscription page in the console under the API Key section.
Important information before subscribing
- Usage scope: This plan is intended for interactive use only within coding tools and agent tools such as Claude Code, Cursor, Qwen Code, Qoder, Qoder CN, and OpenClaw. It must not be used for automation scripts, custom application backends, or any non-interactive batch call scenarios. Models are accessed through the tools' extension mechanisms (Skill, Slash Command, or Agent); calls initiated interactively within a tool are considered normal use. Using the plan API Key for calls outside the allowed scope will be considered a violation or abuse, and may result in subscription suspension or API Key banning.
- Region and data transfer: For Token Plan, the only available region is Singapore and the only available deployment mode is Global, which means the model inference will be performed globally. The prompts and model outputs will involve cross-border data transfer, and you are responsible for ensuring that your use of the Token Plan complies with applicable laws and regulations.
- Account usage policy: The plan is for the exclusive use of the subscriber and sharing is prohibited. Account sharing may result in restricted subscription benefits.
- Purchase limit: Each verified identity is limited to one subscription per edition, and can hold the Personal Edition and Team Edition at the same time.
- Device usage: Token Plan Personal Edition is intended for use on a single device per individual user. Sharing the API Key with others may be considered a violation and result in suspension.
- RAM user identity verification: RAM users do not need to complete identity verification separately. As long as the Alibaba Cloud account has completed identity verification and has assigned the RAM user a seat and API call permission, the RAM user can use Token Plan normally.