SiliconFlow
Fast and affordable AI inference platform. 2.3x faster speeds and 32% lower latency than major cloud platforms. Supports LLM, image, video, and audio models.
At a glance: SiliconFlow has 24 models listed on Inference Hub: 13 LLMs, 8 image models, 2 video models and 1 coding model. The cheapest text model is DeepSeek V4 Flash at $0.130 input and $0.280 output per 1M tokens, and the most expensive is GLM 5.2 at $1.30 input and $4.09 output per 1M tokens. Image generation starts at $0.0050 per image with Z-Image. Free tier: $1 free credit on signup. Prices are checked against SiliconFlow's own pricing every week.
Features
Compliance
None listed
Pricing Model
per tokenDetails
https://api.siliconflow.cn/v1 Model Catalog (27)
SiliconFlow guides
SiliconFlow API FAQ
Is the SiliconFlow API free?
Yes, SiliconFlow has a free tier: $1 free credit on signup. Paid usage is billed per model.
What is the cheapest SiliconFlow model?
DeepSeek V4 Flash at $0.13 in / $0.28 out per 1M tokens, based on the 27 models we track for SiliconFlow.
How many models does SiliconFlow offer?
We track 27 SiliconFlow models: 16 LLM, 8 image generation, 2 video generation, 1 coding.
Is the SiliconFlow API OpenAI-compatible?
Yes. SiliconFlow exposes an OpenAI-compatible endpoint, so you can use the OpenAI SDKs by changing the base URL and API key.
Run SiliconFlow? Add a "Prices tracked on Inference Hub" badge to your site, or email hello@inferencehub.org if a price here is wrong.