AI search company You.com has introduced Inference Router, a managed API service built for running open-source models in production. The product is compatible with the OpenAI API and can route requests across 12 live models. You.com says the service comes with a 99.9% minimum uptime SLA, a zero data retention policy, 100% U.S. hosting, and server-side web search. It uses a pay-as-you-go pricing model and does not charge a platform fee.
The company said the integrated models include Kimi K3, Qwen3.8 Max, GLM-5.3, and DeepSeek V4 Pro, all hosted in the United States. Pricing examples listed in the report show Kimi K3 at $2.85 per million input tokens and $14.25 per million output tokens, while Qwen3.8 Max is priced at $1.85 per million input tokens and $5.55 per million output tokens.
Requests are first sent to You.com’s own inference infrastructure. If that system goes down or reaches capacity, traffic fails over to vetted partners including Parasail, Wafer, Alibaba Cloud, Meta, and Inception. According to the report, all partners must meet the same zero data retention and U.S. hosting requirements.
AI search company You.com has launched Inference Router, a managed API service designed to run open-source models in production environments.
The service is compatible with the OpenAI API and can route requests to 12 live models. It comes with a 99.9% minimum uptime service-level agreement, a zero data retention policy, 100% U.S. hosting, and server-side web search. The pricing model is usage-based, with no platform fee.
Models and pricing
The models integrated into the service include Kimi K3, Qwen3.8 Max, GLM-5.3, and DeepSeek V4 Pro, all of which are hosted in the United States.
For pricing, Kimi K3 is listed at $2.85 per million input tokens and $14.25 per million output tokens. Qwen3.8 Max is priced at $1.85 per million input tokens and $5.55 per million output tokens.
Routing and failover setup
Requests are first sent to You.com’s own inference infrastructure. If that infrastructure goes down or is fully loaded, traffic fails over to vetted partners, including Parasail, Wafer, Alibaba Cloud, Meta, and Inception.
The report said all partners are required to meet zero data retention and U.S. hosting standards.
The item was cited by MarkTechPost.
This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan. Disclaimer:
The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.
Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.
