Alibaba Model Studio Token Plan: Pricing Structure and Claude Code Integration
Review of Alibaba Model Studio Token Plan terms, Singapore availability, Personal and Team quotas, API compatibility, and billing checks before subscribing.
Published: Last checked:

Review of Alibaba Model Studio Token Plan terms, Singapore availability, Personal and Team quotas, API compatibility, and billing checks before subscribing.
Written and checked by the RankLLMs editorial team. See theeditorial policy.
Alibaba Cloud Model Studio’s Token Plan is a Singapore-region subscription that charges usage in Credits. It is a separate product from Alibaba’s Coding Plan and from normal pay-as-you-go API billing. The precise model allowlist, endpoint, price, and quota depend on the current plan, so check the official console before configuring a client.
This review uses Alibaba’s Token Plan overview, last updated September 10, 2026, plus its integration guide. Prices and product limits can change; this is not a live account quote.
Published Personal Edition terms
The official overview currently lists three Personal tiers and promotional prices:
| Tier | Promo price | Listed 7-day quota | Concurrent agents |
|---|---|---|---|
| Lite | $6/month (regular $8) | 2,500 Credits | 1-2 |
| Standard | $18/month (regular $25) | 10,000 Credits | 3-4 |
| Pro | $68/month (regular $80) | 40,000 Credits | 6-8 |
The same page lists a $15 Extra Bundle with 20,000 Credits. It says an active subscription is required and up to five can be held. Confirm bundle validity and deduction behavior in the current purchase terms before buying; product terms can differ between editions.
The overview says the Personal quota is a seven-day fixed window beginning at first use. Once cumulative consumption reaches the tier limit, calls pause until the full window elapses. Unused quota does not carry forward. That is not a continuously rolling calculation where each old request frees credits exactly seven days later.
Team plans and region
Alibaba says Token Plan is currently available only in the Singapore region. The Team edition is a separate product. Its overview describes seat-based monthly quotas and centralized team management; Team standard seats show 25,000 Credits per seat per month, Pro 100,000, and Max 250,000. Team quotas are not the Personal seven-day allowance. Check seat prices and terms in the console, as the product page may be updated independently.
The documentation lists compatible integrations with tools such as Claude Code, Cursor, Qwen Code, Codex, Qoder, and OpenClaw, but compatibility depends on API protocol support. A client needs a custom base URL and key and must use the supported OpenAI or Anthropic protocol. A similar-looking interface alone does not make a tool compatible.
Credit use is not API token price
A Credit quota is not directly convertible to a per-million-token API rate without the plan’s model-specific deduction rules. Model choice, thinking mode, tool calls, and cached input can affect consumption. Track Credits in the Model Studio console and compare them with the same work done through the relevant pay-as-you-go API route.
For an apples-to-apples evaluation, run a fixed set of prompts and record model identifier, endpoint, tools, completion quality, Credits consumed, retries, and task completion. Keep code and prompts out of the test if your account terms do not permit sending them. Team and Personal editions have different terms, so review the applicable data provisions before using private repositories.
What to verify before subscribing
- Switch the console to Singapore and confirm that the plan is available for your account.
- Check that the model you want appears in the Token Plan allowlist; unsupported models may use another billing route or fail.
- Use the dedicated Token Plan key and base URL. Alibaba notes that a general Model Studio endpoint or Coding Plan key can send calls to pay-as-you-go billing.
- Compare fixed-window quotas, bundle conditions, renewals, and any team data commitments with the expected workload.
Token Plan may simplify access to multiple supported models and tools through one subscription. Its value depends on the model allowlist, actual Credit deduction, account region, and the client’s API compatibility. Do not equate it with unlimited usage or assume every model in Model Studio is included.
See Alibaba’s official overview and Token Plan guide, and use our coding plans comparison for broader subscription context.
Frequently Asked Questions
What is the Alibaba Model Studio Token Plan?
Token Plan is an AI model subscription from Alibaba Cloud Model Studio. The current official overview lists Personal and Team editions, model and Harness tool access, and OpenAI- or Anthropic-protocol integration for compatible clients.
Why is Token Plan restricted to the Singapore region?
Alibaba’s current overview says Token Plan is available only in the Singapore region. Switch the Model Studio console to Singapore to purchase and use it.
How does the Personal Edition 7-day quota window work?
Personal Edition uses a seven-day fixed window that starts with the first invocation. When the quota is reached, calls pause until the full window elapses; unused credits do not carry over.
What is the 50% night discount on Alibaba Token Plan?
The published Token Plan overview does not specify this night discount. Check the current plan and model billing details in the console before assuming off-peak usage consumes fewer Credits.
Can I use Alibaba Token Plan with Claude Code and Cursor?
Yes, if the client supports a custom base URL and the OpenAI or Anthropic API protocol. Use the Token Plan’s dedicated key and base URL; using a general Model Studio endpoint can route calls to pay-as-you-go billing.
Source links for data and claims discussed in this article are listed below. Refer to each source for its current figures and methodology.
Recommended Reading
Related by topic and recency.
GLM-5.3-Flash API Pricing: Current Rates and Coding Plan Quotas
Current Z.ai GLM-5.3-Flash API rates are $0.15/M input, $0.03/M cached input, and $0.50/M output. See how API billing differs from Coding Plan quota discounts.
Claude Sonnet 5: Permanent $2/$10 Pricing and Developer Impact
Anthropic made Claude Sonnet 5 API pricing permanent at $2 per million input and $10 per million output tokens. See a cost calculation with cache assumptions.
Claude Code vs Gemini CLI vs GitHub Copilot CLI (2026 Tested)
Terminal AI coding agents compared: Claude Code CLI, Google Gemini CLI, and GitHub Copilot CLI tested on multi-file edits, context limits, and pricing.
