Find a model, connect your tools and follow your usage. Make space for the work itself.
One account. One workspace.
More room to focus.
RAIN FLOW / WORKSPACEProduct preview
MODEL LIBRARY
Different models. New possibilities.
◈
Choose GLM, DeepSeek, Kimi, MiMo or Qwen for code, reasoning and everyday work. Explore currently available models, capabilities and prices in the catalog.
All four plans offer optional wallet fallback. Each 5-hour or weekly window starts with its first request and resets in full when it ends. Check the console for current prices, limits and available models.
Just what you need, when you need it.
Top up your wallet, then pay in account credits for the tokens you use, with no subscription. Model prices are listed in credits per million tokens, ideal for occasional or variable workloads.
Review requests and balance changes in usage and wallet history. Top-ups convert to credits at the rate shown on the payment page.
A FEW GOOD QUESTIONS
Before you begin.
Which tools can I connect?+
Tools with custom API URLs and keys that support OpenAI or Anthropic formats can connect, including Claude Code and Cline. OpenAI SDKs use /v1 or /coding/v1; Claude Code uses the site root or /coding. Find details in the console.
Should I subscribe or pay as I go?+
Choose Coding Plan for ongoing development, or pay as you go for occasional use. Plans are charged in wallet credits every 30 days. Renewals extend the paid term; upgrades credit unused time and start a new cycle. Direct downgrades are not currently supported.
How do free models work?+
The API channel requires at least ¥1 in lifetime top-ups to access free models. Coding users should use /coding and follow their plan’s model permissions. Daily quotas count input, output and cache-write tokens across both channels and reset at 04:00 Beijing time (UTC+8). Exhausted quotas return 429; account and key limits also apply.
Do I need an account for each model?+
No. Use one RainFlow TB account to explore models, create keys and track usage. Available models depend on your account or subscription permissions.
MAKE SOMETHING THAT MATTERS.
Let your next idea take shape.
A little room to think. A simple connection to begin.
Last updated: August 28, 2026. Welcome to the RainFlow TB Open Platform (“we”, “us”). By using our large-model API services (the web console and API endpoints, collectively the “Service”), you acknowledge that you have read, understood and agree to be bound by these Terms.
1. The Service
The Service provides API access to large language models under two billing modes: pay-as-you-go (wallet balance) and subscription (Coding Plans). We continuously improve the Service and may adjust features, model availability and plan benefits from time to time; material changes will be announced on the site.
2. Account and Registration
1. You must provide accurate registration information and keep your credentials (password, API keys, sessions) secure. You are responsible for all activity under your account.
2. Accounts are for personal use only and must not be sold, rented, shared or transferred to any third party.
3. If you suspect unauthorized access or a security issue, notify us immediately via the support channel.
3. Subscriptions, Billing and Refunds
1. Plan purchases, renewals and upgrades are charged from your wallet balance. Renewals append a new cycle after your current paid period; upgrades are prorated against the remaining time of the current cycle, and used quota is carried over to the new plan in proportion to the remaining time.
2. Plan quota includes the cycle quota plus 5-hour and weekly window limits. After the quota is exhausted you may choose to keep billing from your wallet balance (can be disabled in the console).
3. Top-up balance and purchased plans are virtual services. Once purchased, they are non-refundable and non-withdrawable, except where required by law.
4. Free Models
Free models carry a per-user daily quota. We may adjust the scope, quota and availability of free models based on operating costs, with notice on the site.
5. Acceptable Use
You agree not to use the Service to:
1. Share your account or plan quota with third parties, or resell the Service in any form;
2. Abuse free models with scripts or similar means, or otherwise disrupt the Service;
3. Generate or distribute content that violates applicable laws, or infringes others’ intellectual property or personal information rights;
4. Engage in any other conduct that violates laws, these Terms or public order and morals.
Violations may result in warnings, feature restrictions, suspension or termination of your account, without refund of fees already paid.
6. Intellectual Property
The software, interface, trademarks and documentation of the Service are owned by us. Rights in content generated through the Service are governed by applicable law; you are responsible for ensuring your inputs do not infringe third-party rights.
7. Availability and Disclaimers
1. The Service is provided “as is”. Model output is generated by AI and may be incorrect or incomplete; verify it before use. We are not liable for the accuracy or completeness of such output.
2. We are not liable for interruptions or changes caused by force majeure, network failures, upstream model provider adjustments or maintenance, but will endeavor to notify you in advance and restore service promptly.
8. Governing Law and Disputes
These Terms are governed by the laws of the People’s Republic of China. Disputes arising from these Terms should be resolved through friendly negotiation; failing that, either party may bring the dispute to the competent people’s court at the location of the Service operator.
9. Changes to These Terms
We may revise these Terms from time to time; material changes will be announced on the site. Continued use of the Service after changes constitutes acceptance of the revised Terms.
10. Contact Us
Questions about these Terms? Email ceo@rainflowtb.com or reach us via the in-console “Feedback” feature.
Last updated: August 28, 2026. This policy is issued by the RainFlow TB Open Platform (“we”, “us”) in accordance with the Personal Information Protection Law of the People’s Republic of China, the Cybersecurity Law and the Data Security Law. It explains how we collect, use, store and protect your personal information when you use the Service (web console and API endpoints), and the rights you are entitled to by law.
1. Information We Collect
Following the principle of minimum necessity, we collect only what is required to provide the Service:
1. Account information: username, display name and password you provide at registration (stored only in irreversibly hashed form); when you sign in with a third-party account, we receive the user identifier and nickname returned by that platform.
2. Usage data: API call records (model used, token usage, fees, request time, IP address, User-Agent).
3. Transaction information: top-ups, plan purchases, order records and wallet transactions.
4. Content you submit: feedback you provide, and prompts and conversation content you send through the API.
2. How We Use the Information
1. To provide, maintain and improve the Service (authentication, billing and settlement, usage statistics);
2. For security and abuse prevention: detecting misuse and account theft to protect the Service and your account;
3. To comply with legal obligations, and to contact you when necessary (service notices, plan expiry reminders, announcements).
3. Sharing and Entrusted Processing
We do not sell your personal information. We share it only where necessary:
1. Upstream model providers: to fulfill the inference requests you initiate, we transmit the conversation content of that request to the relevant model provider, which does not include your direct identifiers such as account or password;
2. Payment and sign-in providers: the order information necessary to interact with the payment channel when topping up; the authorization flow with the corresponding platform when you sign in with a third-party account;
3. Disclosures required by laws, judicial authorities or administrative agencies.
4. Storage and Security
Your personal information is stored on the servers that run the Service; we do not provide it to parties outside mainland China. We protect it with encrypted storage, access controls and least-privilege authorization. We retain the information for as long as your account exists; after account deletion, we will delete or anonymize it within a reasonable period, except where the law requires otherwise.
5. Local Storage
We use localStorage in your browser to store login credentials, interface preferences and language settings, in order to keep you signed in and remember your settings. You can remove them at any time by clearing your browser data.
6. Your Rights
Under the Personal Information Protection Law, you have the right to access, copy and correct your personal information, to delete it and deactivate your account, and to withdraw consent. Send requests to ceo@rainflowtb.com or via the in-console “Feedback” feature; we will respond within fifteen business days.
7. Protection of Minors
The Service is not directed to children under the age of fourteen. If you are under fourteen, please use the Service only with the accompaniment and consent of a guardian.
8. Policy Updates
We may revise this policy from time to time; material changes will be announced on the site. Continued use of the Service after an update constitutes acceptance of the revised policy.
9. Contact Us
Questions about this policy or personal information protection? Email ceo@rainflowtb.com.
Contact
Questions, business inquiries or any suggestions — drop us a line:
1. In-console feedback: submit issues with screenshots on the “Feedback” page after signing in, with trackable progress;
2. Documentation: the console’s “Connect → Reference docs” page covers protocols, error codes and plan rules.
FAQ
1. Plan quota, renewal and upgrade rules: see “Connect → Reference docs → Plan rules” in the console;
2. API errors: see “Connect → Reference docs → Error codes”;
3. Free models: the API channel requires at least ¥1 in lifetime top-ups. Quotas count input, output and cache-write tokens across both channels and reset at 04:00 Beijing time (UTC+8).