- vendor
- Featherless AI
- whatItIs
- Serverless LLM inference platform giving developers one API key to access 30,000+ open-weight models (Qwen, Llama, Mistral, DeepSeek, RWKV) on flat-rate monthly subscriptions with unlimited tokens.
- hosting
- managed-only
- pricingModel
- flat-rate subscription ($10/mo Basic, $25/mo Premium, $100/mo Agent Standard)
- modelProviders
- open-weight (30,000+ models from Hugging Face)
- stateModel
- stateless
- toolModel
- OpenAI-compatible API
- includesBrowser
- false
- includesMemory
- false
- longRunning
- per-request
- hitlSupport
- false
- observability
- basic logs
- maturity
- ga
- launched
- 2024
- notes
- Flat-rate beats per-token pricing 1.5–2x at 10M tokens/day. Largest Hugging Face inference partner with 6,700+ curated models. Built by RWKV Linux Foundation contributors.
