Featherless

Featherless is a serverless large language model inference platform that provides access to over 40,000 open-source models through a single API. The platform enables users to deploy, fine-tune, test, and run production AI workloads without setup or hosting overhead, offering low-latency, dependable uptime with architecture designed for real workloads. Featherless offers tiered pricing including a Chat plan at $25/month with unlimited tokens, a Developer plan at $50/month with 256K context and fastest response times, and a Business plan with dedicated GPUs including H100, MI325, B200, and B300 configurations. The platform supports diverse model architectures and provides burst and failover to public cloud. Built by researchers contributing to RWKV, a Linux Foundation project, Featherless hosts trending models from Hugging Face including DeepSeek, GLM, Qwen, Llama, and GPT-OSS. Founded in 2023 and headquartered in San Francisco, California, Featherless has raised $25.0M in Series A funding from investors including Airbus Ventures, 500 Global, AMD, BMW i Ventures, Kickstart Ventures, Panache Ventures, and Wavemaker Partners.