SiliconFlow
Hosted inference API for open and commercial LLM and multimodal models
Linked from siliconflow.com · checked 29 Oct
- Tool type
- Hosted AI inference API platform
- Made for
- Developers and technical teams
- Runs on
- Cloud-hosted API and GPU infrastructure
- Models are called over API; API keys and the cloud console are used for setup, not as the runtime.
What it does
SiliconFlow is a hosted AI inference platform that gives developers one API for open and commercial language models and for image, video, and audio models. Requests go in over the API and come back as chat completions, generated images, videos, or speech output.
Teams use it to add model calls to applications such as coding assistants, retrieval-augmented generation, agent workflows, content generation, and search. Deployment options include serverless pay-per-use access, fine-tuning with one-click deployment, reserved GPUs, elastic GPUs, and an AI Gateway with routing, rate limits, and cost control. The site describes the API as OpenAI-compatible.
It suits developers and technical teams who want usage-based billing. Chat models are billed per million input and output tokens, while image, video, and audio models are priced by output. Setup means creating an account and getting an API key. Volume discounts and enterprise terms go through the sales team.
What it can do
- Serverless access to a catalog of open and commercial LLMs via one API call
- Text, image, video, and audio model inference
- Fine-tuning of supported models with one-click deployment
- Reserved and elastic GPU deployment options
- AI Gateway with smart routing, rate limits, and cost control
- OpenAI-compatible API surface
Before you choose
- Per-model and per-task rates differ widely; check the pricing page for the exact model before estimating cost.
- Reserved GPU and enterprise pricing are not published on the pricing page and require contacting sales.
- Model catalog changes frequently with new releases; verify a model is listed before building against it.
Listed under
Get started
Free credits are mentioned on the pricing page; confirm current terms in the console.
- Create an account at the cloud console.
- Generate an API key from your account.
- Call a chosen model through the OpenAI-compatible API, using the docs for code examples.
Plans and prices
Prices in USDPay-as-you-go with no minimum commitments. The pricing FAQ says new accounts can start with $1 in free credits. Monthly spending limits can be set in the account dashboard. Volume discounts are available through sales. Rates vary by model and change with releases.
- No minimum commitments
- Monthly spending limits in dashboard
Source: siliconflow.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
- SiliconFlow – AI Infrastructure for LLMs & Multimodal Models ↗www.siliconflow.com
- Pricing Plans - SiliconFlow | Transparent Pay-as-You-Go ↗www.siliconflow.com
- SiliconFlow – AI Infrastructure for LLMs & Multimodal Models ↗www.siliconflow.com
- About Us - SiliconFlow | Global AI Infrastructure Provider ↗www.siliconflow.com
- Contact Us - SiliconFlow | Sales & Technical Support ↗www.siliconflow.com
Ratings and reviews
No ratings for SiliconFlow yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- siliconflow.com: product, pricing and policy pages
- YC directory mirror
- Batch and company name
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
- manual_review
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.