Ultra-Fast Open Model Cloud Inference
SiliconFlow (SiliconCloud) delivers scalable model inference optimized with high-efficiency kernel acceleration. Upon signing up, developers receive an allocation of 20 million free tokens valid across standard open-weight architectures including DeepSeek-V3, DeepSeek-R1, and Qwen 2.5.
Included Free Token Allocation
- **Frontier Reasoning**: Full access to `deepseek-ai/DeepSeek-R1` and `deepseek-ai/DeepSeek-V3`.
- **Coding & Multimodal**: Fast inference on `Qwen/Qwen2.5-Coder-32B-Instruct` and `telechat2`.
- **OpenAI Compatible Endpoint**: Simple drop-in integration with standard OpenAI SDKs via `https://api.siliconflow.cn/v1`.
Claim Instructions
1. Visit the SiliconFlow Cloud registration page at cloud.siliconflow.cn.
2. Create an account via GitHub or email verification.
3. Go to the API Keys section to copy your key and access your 20M token balance.