Featherless.ai

Run long-context open-weight models in Cline with Featherless.

$25/month unlimited plan for AI research, coding, writing, and chat. Trusted by YouTube, Cisco, BlackRock & more.

Unlimited AI models and tools for writing, research, learning, and creativity. $25/month for everything. No limits, no surprises.

Dedicated nodes for high-volume, background AI workloads. Process billions of tokens monthly with guaranteed savings or full refund.

A base model refuses 43.8% of system-hardening prompts. What abliteration buys on the benchmarks, and every refusal-modified model Featherless serves.

Seven open-weight models for fiction and long-form prose, scored on the creative-writing benchmarks that exist, plus the context Featherless serves each at.

Featherless has raised a $20 million Series A round to accelerate one mission: making open-source AI practical, sovereign, reliable and effortless at scale.

All 40,000+ open models on every plan. What changes are context, concurrency, and intended use.

Featherless compares its flat-rate unlimited inference pricing with OpenAI, Anthropic, Google, and other providers to estimate potential cost savings.

A Fireworks AI alternative that doesn't retire your model. 40,000+ open models on one OpenAI-compatible API at one rate, with no Priority or Fast tiers, or a managed node from $7,500/mo.

Featherless offers early access to OpenAI’s new open-source reasoning model with zero-click deployment and an OpenAI-compatible API.

Replicate's open-LLM shelf stops at DeepSeek V3.1 and Kimi K2.6. Featherless serves Kimi K3, DeepSeek V4, GLM 5.3, Gemma 4 and 40,000+ open models on one OpenAI-compatible API, billed per token.

Featherless provides serverless AI inference access to over 40,000 open-source models via a single API. The company highlights its team, research background, and $20 million Series A funding.

Freedom to reliably deploy any open model effortlessly.