Quick setup
Open the kRouter dashboard at http://localhost:20128
Go to Providers → click SiliconFlow
Paste your API key, click Save
Test the connection
Use in your IDE
Point any OpenAI-compatible client at kRouter using your local API key.
# Endpoint
http://localhost:20128/v1
# API key (from Dashboard → API Keys)
sk-krouter-XXXX
# Model
siliconflow/deepseek-v3How SiliconFlow billing works with kRouter
SiliconFlow is pay-per-token. You bring your own API key and kRouter routes to it, so you keep your own rate limits and billing relationship. kRouter authenticates it via apikey. 16 models are routable as siliconflow/<model>.
Through kRouter, SiliconFlow serves chat and code completion and image generation, exposed on /v1/images/generations. Requests reach it through the same local OpenAI-compatible endpoint as every other provider, so switching to or from SiliconFlow is a one-line model change in your client rather than a rewrite.
SiliconFlow models & pricing
Prices are USD per 1M tokens, cheapest first. kRouter bills nothing on top; this is the provider's own rate.
| Model | Route as | Input /1M | Output /1M |
|---|---|---|---|
| DeepSeek V4 Pro | siliconflow/deepseek-ai/DeepSeek-V4-Pro | $0.14 | $0.28 |
| DeepSeek V4 Flash | siliconflow/deepseek-ai/DeepSeek-V4-Flash | $0.14 | $0.28 |
| DeepSeek V3.2 | siliconflow/deepseek-ai/DeepSeek-V3.2 | $0.14 | $0.28 |
| DeepSeek V3.2 Exp | siliconflow/deepseek-ai/DeepSeek-V3.2-Exp | $0.14 | $0.28 |
| DeepSeek V3.1 | siliconflow/deepseek-ai/DeepSeek-V3.1 | $0.14 | $0.28 |
| DeepSeek V3.1 Terminus | siliconflow/deepseek-ai/DeepSeek-V3.1-Terminus | $0.14 | $0.28 |
| DeepSeek R1 | siliconflow/deepseek-ai/DeepSeek-R1 | $0.14 | $0.28 |
| Qwen 3.5 397B A17B | siliconflow/Qwen/Qwen3.5-397B-A17B | $0.50 | $2.00 |
| Qwen 3.5 122B A10B | siliconflow/Qwen/Qwen3.5-122B-A10B | $0.50 | $2.00 |
| GLM 5.1 | siliconflow/zai-org/GLM-5.1 | — | — |
| GLM 5 | siliconflow/zai-org/GLM-5 | — | — |
| Kimi K2.6 | siliconflow/moonshotai/Kimi-K2.6 | — | — |
| Kimi K2.5 | siliconflow/moonshotai/Kimi-K2.5 | — | — |
| GPT OSS 120B | siliconflow/openai/gpt-oss-120b | — | — |
| MiniMax M2.5 | siliconflow/MiniMaxAI/MiniMax-M2.5 | — | — |
| Ling Flash 2.0 | siliconflow/inclusionAI/Ling-flash-2.0 | — | — |
SiliconFlow links
SiliconFlow FAQ
Is SiliconFlow free through kRouter?
SiliconFlow is pay-per-token: you bring your own API key and are billed by SiliconFlow at their rates. kRouter adds nothing on top and is free and MIT-licensed.
How do I use SiliconFlow with Claude Code or Cursor?
Connect SiliconFlow in the kRouter dashboard, then point your tool at kRouter's local endpoint (http://localhost:20128/v1) with a kRouter API key. Any OpenAI-compatible client works, and models are addressed as siliconflow/<model>.
Which SiliconFlow models can kRouter route to?
16 models, including DeepSeek V4 Pro, DeepSeek V4 Flash, DeepSeek V3.2. Every one is addressed as siliconflow/<model> from any client.
What happens when SiliconFlow is rate limited?
kRouter detects the limit, parks that account until its real reset time, and fails over to your next connected account or provider — so the request still completes instead of erroring.