Products
API Relay
OpenAI-compatible protocol
Smart Routing
Multi-channel auto-switching
Enterprise Security
IP whitelist & rate limiting
Pay-as-you-go
No minimum spend
| Model | Feature | Input / 1M Token | Output / 1M Token |
|---|---|---|---|
| Claude Opus 4.7 | Best for Coding | ¥35 | ¥175 |
| Claude Opus 4.6 | Deep Reasoning | ¥35 | ¥175 |
| Claude Sonnet 4.6 | Well Balanced | ¥21 | ¥105 |
| Claude Sonnet 4.5 | Stable & Reliable | ¥21 | ¥105 |
| Claude Haiku 4.5 | Fast & Light | ¥7 | ¥35 |
| DeepSeek v3.2 | Best Value | ¥4.34 | ¥12.95 |
| MiniMax M2.5 | Multimodal | ¥2.1 | ¥8.4 |
| Kimi K2 Thinking | Deep Thinking | ¥4.2 | ¥17.5 |
| Kimi K2.5 | Long Context | ¥4.2 | ¥21 |
| Qwen3 32B | Open Source | ¥1.4 | ¥5.46 |
| Qwen3 Coder Next | Coding Pro | ¥3.5 | ¥8.4 |
Free
¥0
¥100 credit after approval
Recommended
Pay-as-you-go
Top up & use
No minimum spend
Enterprise
Contact Us
Dedicated channel · SLA
Documentation
Direct links to provider documentation
🅰
Anthropic
Claude family for coding and reasoning
Official DocsView →
🔍
DeepSeek
High value-for-money LLMs
Official DocsView →
M
MiniMax
Chinese multimodal LLMs
Official DocsView →
🌙
Moonshot
Kimi family with long context
Official DocsView →
Q
Alibaba
Qwen open source family
Official DocsView →
Quick Start
client = OpenAI(
api_key="sk-zx-...",
base_url="https://api.revelry.ai/v1"
)
response = client.chat.completions.create(
model="claude-opus-4-7",
messages=[{...}]
)