SiliconFlow & Fast Inference Gateways Guide

For creators seeking lightning-fast inference or direct high-availability routes, high-throughput inference gateways provide seamless access to DeepSeek-R1, DeepSeek-V3, and open-source models.

---

1. SiliconFlow (Fast Domestic & Global Access)

SiliconFlow delivers optimized inference for top open-source models.

• Top Models: deepseek-ai/DeepSeek-R1, Pro/deepseek-ai/DeepSeek-R1, deepseek-ai/DeepSeek-V3, Qwen/Qwen2.5-72B-Instruct;
• Setup: In DuRT, add SiliconFlow under AI & LLM, paste your key, with Base URL https://api.siliconflow.cn/v1.

---

2. Groq (Ultra-Low Latency LPU Engine)

Groq uses custom LPU (Language Processing Unit) silicon, generating hundreds of tokens per second.

• Models: llama-3.3-70b-versatile, llama-3.1-8b-instant;
• Setup: Base URL https://api.groq.com/openai/v1.

---

3. OpenRouter (Unified Model Routing)

OpenRouter aggregates hundreds of models with dynamic failover routing.

• Setup: Base URL https://openrouter.ai/api/v1.