SiliconFlow
SiliconFlow serves open and commercial language, image and video models through serverless per-token APIs, with fine-tuning and dedicated deployments of public or custom models.
Find alternatives to SiliconFlow
What it does
- Model inference
- Model training
Features 5
- Serverless inferenceCalls hosted models without setup, paid per use, with automatic scaling for traffic spikes.
- Managed fine-tuning pipelineCustomizes models on the user's data through a fully managed pipeline, in three steps.
- Reserved GPU capacityDedicated, always-on GPU compute for long-running and mission-critical workloads.
- AI gateway with rate limitsUnified model access with smart routing, rate limits and cost control.
- OpenAI-compatible APIOne API covers all hosted models and is fully OpenAI-compatible.
Sources
- Pricing
-
Products page read 2026-11-16: on-demand billing ("Pay only for what you use") and Reserved GPUs ("Lock in consistent capacity for long-running jobs with significant cost savings compared to on-demand pricing.") whose only route is "Contact Sales", so price on request is a second model; the serverless per-token prices are on the pricing page named in pricing_url.
- Description
-
Official documentation · 10 Nov 2026
Product page: "A one-stop AI platform for inference, fine-tuning, and custom deployment—flexible, scalable, and developer-friendly."
- What it does
-
Official documentation · 10 Nov 2026
Product page: "A one-stop AI platform for inference, fine-tuning, and custom deployment—flexible, scalable, and developer-friendly."
- Sold within
-
Sold on its own with published plans.
- Name
-
Official documentation · 10 Nov 2026
Name as published on the product page.
- URL
-
Official documentation · 10 Nov 2026
Product page, read 2026-11-10.
- Deployment
-
Official documentation · 10 Nov 2026
Product page: "Instantly call powerful models without setup."
- Interfaces
-
Official documentation · 10 Nov 2026
Product page: "Instantly call powerful models without setup."
Open SiliconFlow in the directory
Something wrong here? Send a correction — quote this product id: siliconflow-platform.