Full Large Model Support and Channel Capability Matrix
This page summarizes the main model ecosystems, channel tiers, and recommended group identifiers currently offered by APIShop, making it easier to evaluate available models and channel routing strategies before integration.
1. Our Core Resource Advantages
- Unified ecosystem access: integrate once to seamlessly call top-tier model families from OpenAI, Anthropic, Google, DeepSeek, and other leading providers worldwide.
- Native-grade SLA assurance: core channels connect directly to leading compliant cloud providers such as AWS and GCP, providing a solid foundation for mission-critical production environments.
- Exclusive high-concurrency pool: to address the strict Rate Limit constraints of official APIs, the platform exclusively maintains high-quality multi-node capacity and dedicated AZ nodes to handle enterprise-scale concurrency spikes in the tens of millions.
- Intelligent routing and scheduling: provides channel configurations at different service tiers, allowing enterprises to switch freely between “stability first” and “price first” based on the importance of each workload.
2. Core Large Model Support Matrix
Below are the core model ecosystems supported by the platform and the recommended routing group identifiers. The system will dynamically route to the best-matched physical channel based on your business needs.
🌐 International Core Large Language Models
| Ecosystem | Channel Service Tier | Resource Characteristics and SLA Assurance | Recommended Group Identifier |
|---|---|---|---|
| OpenAI (GPT Series) | Direct from Vendor | Official native resources with extremely high SLA, the top choice for production environments. | gpt-of / openai-of |
| Enterprise High-Concurrency Pool | High-quality AZ node pool that ensures strong stability under high concurrency. | gpt-vip | |
| Standard Value Pool | Hybrid resource scheduling with excellent cost performance, suitable for daily development and testing. | gpt-sp | |
| Anthropic (Claude) | Direct from Vendor | Native AWS/Anthropic resources with low latency and zero caching. | cl-of / claude-of |
| Enterprise High-Concurrency Pool | Exclusive large-volume discounted resource pool with a high-cache availability pool. | aws-vip / az-vip | |
| Standard Value Pool | Hybrid scheduled resources, recommended for moderate use. | cl-aws / cl-kr | |
| Google (Gemini) | Direct from Vendor | Native GCP resources with reliable support for ultra-large context processing. | gm-of / veo-of |
| Enterprise High-Concurrency Pool | High-quality large concurrency pool built to withstand heavy loads. | gm-vip | |
| Standard Value Pool | Standard discounted pool with strong cost-performance value. | gm-sp / gm-cli |
Top Large Language Models
| Brand Ecosystem | Channel Service Tier | Resource Characteristics and SLA Coverage | Recommended Group Identifier |
|---|---|---|---|
| DeepSeek | Direct Official Connection | Direct resources connected to the official DeepSeek infrastructure. | ds-of |
| Major Cloud Provider Channel | Highly reliable scheduling powered by official channels from Alibaba Cloud/Volcengine. | al-of / db-of | |
| High-Concurrency and Special-Offer Pool | Supports high-concurrency scheduling with filter removal, along with lower-tier discounted resource pools. | al-vip / db-sp |
(Note: Only some core models are shown above. For the full list of supported models and the latest pricing, please refer to what is actually displayed in the console.)
3. Enterprise-Grade Customization and High-Concurrency Support
For business scenarios with exceptionally high daily call volumes or very strict compliance and stability requirements, our business and architect teams provide dedicated 1-on-1 customized support services:
- Dedicated 10M+ TPM capacity assurance: We provide an exclusive high-concurrency resource channel to effectively handle large-scale, production-grade traffic bursts.
- Exclusive enterprise service system: Enjoy tiered, highly competitive commercial pricing discounts, along with 24/7 high-priority technical support. To obtain the full model pricing list or conduct high-concurrency stress testing for your business, please contact your dedicated account manager.