Inference API·China
SiliconFlow
Chinese inference aggregator — hosts DeepSeek / Qwen / GLM / LLaMA behind one API
Some links are affiliate links. We may earn a commission at no cost to you.
Founded
2023
Headquarters
China
Price range
—
API format
openai
Reachability & verification
Live reachability
Target (SiliconFlow): https://siliconflow.cn
Official Path Verified
ModelHub proactively confirms this provider's identity — it is the official lab/API, not a wrapper or reseller. Reachability is probed live from the gateway egress; overseas reference latencies are estimates.
About SiliconFlow
SiliconFlow (硅基流动) runs an inference aggregator that hosts open-weight Chinese + global models (DeepSeek, Qwen, GLM, LLaMA, FLUX for images). Unified billing, OpenAI-compatible endpoints, and sometimes cheaper than first-party pricing for the same open-weight model.
Pros
- Overseas node available — accessible from outside mainland China
- OpenAI-compatible API: openai — drop-in for existing SDK users
- Established 3+ years (founded 2023)
Cons
- Payments only via alipay, wechat, card — may block non-China billing
Tracked models
0Map a verified path
We've prefilled SiliconFlow's tracked models. Run the planner to see which need brokerage vs. an official overseas route.
Paste the model names your team wants — one per line or comma-separated. We'll map each to a verified path.
Rule-based matching, not ML.
Request a brokerage intro
Not on our verified catalog yet? We broker a vetted intro to SiliconFlow and handle residency, contracting and integration.
Affiliate disclosureSome provider website links may earn us a referral fee. Our verification and brokerage are independent — we only earn when we actually deliver a vetted intro.