◇ PRICING · Transparent / Flexible / Pay-as-you-go

Pay-as-You-Go · No Minimum Spend

Start with Free for personal use, switch to Pro for team/business volume, large enterprises go with Enterprise for private deployment. Model pricing transparent and synced with official vendors, no platform markup.

FREE · Free Tier

Free

Personal trial · Daily free quota

¥ 0/ forever
No credit card required · Sign up and use
Includes
  • Playground chat
  • AI App Marketplace (free apps)
  • Selected free models · daily quota
  • Personal account · community support
MOST POPULAR
PRO · Pay-as-You-Go

Pro

Developers / teams / SaaS · billed by actual token usage.

Pay-as-You-Go
Model prices synced with official vendors · no platform markup
All of FREE +
  • 200+ models all available
  • Unified API · OpenAI SDK compatible
  • Workflows · Agents · Knowledge Base
  • Semantic caching · auto-fallback
  • Team collaboration · up to 20 members
  • Email support within 1 business day
ENTERPRISE · Custom

Enterprise

Large enterprises · private deployment · compliance requirements.

Contact Sales
Contract billing · priced by scale & compliance level
All of PRO +
  • Unlimited members · three-level org
  • SSO · SAML · SCIM · audit
  • Private VPC / on-premises deployment
  • SOC 2 · ISO 27001 · Level 3 Protection
  • Volume tier discounts · up to 40% off
  • 99.99% SLA · dedicated CSM

Model pricing transparent at official rates · cache hits cost 0 tokens · higher volume = deeper discounts

Feature Comparison

FeatureFreeProEnterprise
Scope
Playground Chat
AI App MarketplaceSelected free appsAll appsAll + Custom
Unified API Access
Workflow Orchestration
Enterprise Custom Agents
Knowledge Base / Vector Retrieval
Models & Quotas
Available ModelsSelected free models200+ all200+ all
Daily Free QuotaFree model quota
Balance Top-upPay per tokenContract billing
QPS Limit5200Custom
Semantic Caching
Multi-Model Fallback
Team & Enterprise
Account TypePersonalPersonal / TeamEnterprise Org
Members1Up to 20Unlimited
Three-Level Org Structure
Employee Token QuotaBasicFine-grained
SSO · SAML · SCIM
Audit Trail · SIEM Export
Auto PII Masking
Deployment & Compliance
SaaS Shared
Private VPC Deployment
SOC 2 · ISO 27001 · Level 3
No Retention / No Training AgreementOn by defaultContract explicit
Support
Response TimeCommunity1 business day4-hour SLA
SLA99%99.1%99.9%
Dedicated CSM
Compliance Audit Support · DPA · BAA

FAQ

What models are available on Free? How much daily quota?+

Free tier activates selected free models (covering mainstream lightweight models for Chinese / English / code scenarios), with a fixed daily token quota refreshing at 00:00 (UTC+8). Usage is limited to Playground chat and AI apps tagged 'free'. For API calls, custom workflows, enterprise agents, or premium models like Claude / GPT-5 / Gemini, upgrade to Pro.

How is Pro billed? Is there a minimum spend?+

Pro charges by actual token usage with no monthly fee and no minimum spend. After topping up, charges are deducted from your account balance. Pricing is transparent and synced with official vendors — the platform doesn't mark up model costs. Platform revenue comes from value-added services like caching, fallback, and usage sharing. You can pause anytime, and unused balance is refundable.

Does WANFLOW mark up model costs?+

No markup. All model call prices are transparent and synced with official vendors (Anthropic, OpenAI, Google, Alibaba, DeepSeek, etc.). Semantic cache hits cost 0 tokens. Multi-model fallback, key isolation, usage sharing, and other platform capabilities are included at no extra charge. Enterprise customers get deeper volume tier discounts, down to 60% of official pricing.

Will my data be used for training?+

No. WANFLOW has signed no-training agreements with all upstream model vendors. API call prompts/responses will not be used for model training. Pro users have 'no retention' enabled by default, Enterprise can specify data residency, deletion timelines, and cross-border compliance terms in their contract.

Will domestic users calling overseas models (Claude / GPT) be throttled?+

No. WANFLOW has independent nodes in Hong Kong, Singapore, Tokyo, Frankfurt, London, etc., with self-built BGP + Anycast networks. Domestic users route through Shenzhen → Hong Kong → overseas, end-to-end P50 latency 178ms. Overseas model calls go through WANFLOW's cross-border compliance channel, no local compliance issues triggered.

Do you support invoices? Do you support corporate payments?+

Yes. VAT special invoices at 6% tax rate. Pro supports WeChat / Alipay / credit card / corporate account top-up. Enterprise supports annual contracts, quarterly billing, and cross-border USD settlement.

How to upgrade from Pro to Enterprise?+

Contact our enterprise advisor at enterprise@wanflow.ai. We'll assess your team size, compliance requirements, and deployment preferences, providing a PoC + pricing plan within 24 hours. Existing data can be seamlessly migrated, and Pro balance carries over.