XuanShu API enterprise sales and onboarding
Turn your team’s AI access into clear infrastructure
Enterprise and technical team support for API key governance, resilient routing and fault tolerance, subscriptions, pay-as-you-go billing, corporate bank transfer, invoicing, and onboarding.
Team usage governance · Resilient routing and fault tolerance · Corporate bank transfer and invoicing
Unified API access
Compatible with leading AI API protocols
By project and member
Assign quotas, expiration, and IP boundaries
Usage and cost
Review requests and token details by key
Enterprise billing
Subscription, usage-based, bank transfer, and invoicing
Team usage governance
Create separate API keys for projects, environments, or members. Set total quotas, expiration dates, IP allowlists and blocklists, and rolling spend limits. Review requests, tokens, cost, errors, and latency by key to reconcile usage and maintain clear boundaries.
Multiple-key boundaries
Separate keys by project, environment, or member with their own quotas and expiration.
Access rules
Control call scope with IP allowlists, blocklists, and rolling spend limits.
Auditable usage
Review requests, tokens, cost, errors, and latency by key.
OpenAI-style public endpoints
The public Base URL is enabled; model parameters and capabilities depend on the current console configuration.
Anthropic Messages-style public endpoints
Messages and token counting are public; model and tool support depend on the current console configuration.
Gemini v1beta-style public endpoints
Available actions, models, and streaming capabilities depend on current Models responses and console configuration.
API key total quota
Quota is configured per key; actual balance and usage are shown in the console.
API key expiration
Expiration must be configured by the user; expired keys cannot continue making calls.
IP allowlist and blocklist
Rules match configured IPs and do not include time-of-day access windows.
Rolling spend limits
Five-hour, one-day, and seven-day rolling windows are supported; effective limits depend on key configuration.
Call and cost details
Recorded fields include model, tokens, cost, latency, time to first token, and errors; visibility depends on available call records.
Resilient routing and fault tolerance
Model matching, account and channel status filtering, session affinity, candidate switching, and retry policies select a follow-up path for recoverable errors and reduce the impact of a single account, channel, or short-lived failure.
Results still depend on third-party model services, network conditions, client configuration, and request characteristics. Refer to the console and technical documentation for current models and capabilities.
Model matching
Select a suitable path based on the requested model and currently available capabilities.
Status and sessions
Filter by account and channel status, and maintain session affinity when permitted.
Candidate switching
Switch to a follow-up candidate after retryable errors when one is available.
Model matching
Matching depends on current model mappings and account configuration and does not guarantee permanent model availability.
Account status filtering
Filtering applies to current candidate accounts and cannot eliminate every upstream failure.
Session affinity
Session affinity depends on the request and routing state and does not guarantee a fixed account.
Candidate switching after retryable errors
Switching applies only to retryable errors and available candidates and cannot guarantee every request succeeds.
Enterprise billing support
Subscription
For teams that prefer to plan budget and quota around a package; current plans, quotas, and eligible models are shown in the console.
Pay-as-you-go
For variable workloads or gradual adoption, with costs reviewed against actual API usage.
Corporate bank transfer and invoicing
Enterprise customers can use corporate bank transfer and request invoices; sales support explains the process.
Subscription billing
Plans, quotas, pricing, and eligible models are shown in the current console.
Pay-as-you-go billing
Current unit prices, multipliers, balance, and charges are shown in the console.
Enterprise invoicing
Enterprise customers can use corporate bank transfer and request invoices; contact us by email or Discord.
Onboarding process
- Define the team scenario
- Complete technical validation
- Configure governance controls
- Confirm billing arrangement
Enterprise contact template
Copy this template and send it through any of the contact channels below so enterprise sales can understand your integration scope and billing needs.
Open enterprise consultation template
Hello, we would like to evaluate XuanShu API for enterprise use.
Team size:
Tools / SDKs:
Models and business scenario:
Expected usage:
Planned API key count:
Billing preference: Subscription / Pay-as-you-go / To discuss
Invoicing requirements:
Other technical or procurement questions:Do not send complete API keys, passwords, payment credentials, or sensitive request content.
Scan with your phone to add