Accelerate your organization with unified multi-model API access, team subscriptions, and custom enterprise RAG engineering. High-concurrency reverse proxying with low latency, transparent billing, and zero unauthorized data retention.
| Model / Route | Edge Cluster | Protocol | Latency | Status |
|---|---|---|---|---|
|
Claude 3.5 Sonnet
anthropic/claude-3-5 |
US-West BGP | HTTP/2 SSE | 28ms | ● Operational |
|
GPT-4o / GPT-5
openai/gpt-4o-latest |
Wyoming Edge | gRPC Stream | 32ms | ● Operational |
|
DeepSeek-V3 / R1
deepseek/v3-chat |
APAC Relay | REST API | 36ms | ● Operational |
|
Gemini 1.5 Pro
google/gemini-1.5-pro |
Global Anycast | WebSocket | 41ms | ● Operational |
curl -X POST https://api.chengduweistuo.com/v1/chat/completions \ -H "Authorization: Bearer wst-live-edge-token" \ -H "Content-Type: application/json" \ -d '{ "model": "claude-3-5-sonnet", "messages": [{"role": "user", "content": "Ping"}], "stream": true }'
Tell us the product, team size or project goal.
Review availability, charges, third-party costs and delivery.
Work starts after the written order and payment terms are agreed.
Each service is quoted separately so you can see what is included.
// Enterprise Workspace Provisioning
plan: "team_plus_dedicated"
seats: 25
sso: "SAML 2.0 / Okta"
support: "Direct Business Onboarding"
Product and plan selection, subscription arrangements and setup support.
A quote must identify the supplier and the access you receive. Provider names do not imply a partnership or reseller authorization.
Discuss this service →
// Low-Latency Edge Spec
endpoint: "/v1/chat/completions"
protocol: "HTTP/2 SSE Streaming"
rate_limit: "10,000 RPM / 2M TPM"
load_balance: "Multi-Region Failover"
Discuss model access, application integration and usage-based service requirements.
API credits and app subscriptions are separate products. Supported routes and model availability are confirmed in your quote.
Discuss this service →
// Custom Private Cluster
engine: "vLLM / TensorRT-LLM"
vector_db: "Milvus / Qdrant RAG"
quantization: "FP8 / AWQ-INT4"
compliance: "Air-Gapped Local Cluster"
Custom application development, private knowledge retrieval, AI workflows and model deployment.
Hardware, third-party licenses and recurring infrastructure costs are identified separately from development fees.
Discuss this service →Engineered for extreme reliability, low latency, and zero unauthorized data retention.
High-concurrency BGP edge gateway with dynamic model routing, automatic failover, Redis-backed token bucket rate limiting, and real-time streaming multiplexer. Provides standard OpenAI-compatible interfaces for all major upstream model providers.
Engineered dual-hub topology spanning North America (Wyoming Edge) and Asia-Pacific. Leveraging optimized BGP Anycast routing and encrypted transit to deliver guaranteed P95 latency below 38ms.
Complete local document ingestion and multi-agent reasoning pipeline. Supports air-gapped local deployment, high-dimensional vector databases, and hybrid semantic re-ranking with zero data leakage.
Pricing depends on the product, usage, service term or project scope. Request a written quote for your requirements.
Your quote will show the seller, service description, currency, total charges, any applicable taxes, third-party fees, delivery schedule and cancellation terms. This website does not take payments or activate subscriptions.
Request a QuotePractical details for an informed purchase.
Third-party accounts, availability and usage limits follow the applicable provider terms. Confirm the specific arrangement before purchasing.
For integrations, agree which providers process your data and how credentials and logs are handled before connecting production systems.
A written order identifies the deliverables, timing, dependencies and acceptance process.
Use the contact page for service questions, cancellation or refund requests. Any contracted support hours are stated in your order.
Official product information reviewed September 24, 2026. This directory is not a list of subscriptions or models included in a Weistuo order.
GPT-6 Astra · Sol · Luna
Official source ↗Claude Sonnet 5 · Opus 5.5 · Fable 5.1
Official source ↗V4.1 Flash · V4 Pro (0813)
Official source ↗Gemini 3.8 Flash
Official source ↗V8.2
Official source ↗Gen-4.5
Official source ↗v6 · v6-wild · v6-mini
Official source ↗Pro · Pro+ · Ultra · Teams
Official source ↗Pro · Pro+ · Max · Business · Enterprise
Official source ↗How quotes, accounts and delivery work.
No. Start with an inquiry. Any payment request must follow a written order identifying the seller, the service and the accepted payment method.
No. The product directory links to provider information. Your quote specifies the exact product, supported model, term and usage limits.
This website does not enroll you in recurring billing. Any future recurring arrangement must disclose its amount, frequency and cancellation method and receive your explicit agreement.
See the Refund & Cancellation page. Initial subscriptions have the stated seven-day guarantee; separately purchased consumed API usage and completed custom work have different conditions. Existing agreed rights remain in effect.