- SI Score
- 80/100
- Pricing
- Not confirmed
- Free plan
- Not confirmed
- Sign-up required
- Not confirmed
Ask SI about Featherless
Answers use saved listing facts and recorded sources.Homepage screenshot

What it does
Featherless provides serverless hosting for open language models. One API key gives access to a library of 40,000+ models, browsable by category such as reasoning, small models, roleplay, creative writing, productivity and language-specific models. Plans include a chat plan at $25 per month with up to 32K context and 4 concurrent units, a developer plan with $50 credits per month, up to 256K context and usage-based inference, and a business option with dedicated GPUs. Users can sign up with Google, Hugging Face, GitHub, Discord or an email address. The site links to documentation, a quickstart guide, pricing, a cost calculator and a status page.
Featherless lists a model library covering models from providers such as DeepSeek, Qwen, GLM, Gemma, Llama, Mistral, Kimi and MiniMax, with labels like Trending, Most Popular and Top Reasoning. An API key page is available for signed-in users. The site also hosts Simple Jev, an open source classifier built on open models that turns context into structured decisions; it offers a demo API without login limited to 2k context and 2 RPS, a playground, evaluations, API docs and RFDT fine-tuning of decision logits. Simple Jev models on Featherless start at $0.03 per million input tokens and are described as in beta.