Snapshot
Continue to About ↓- What it does
- Featherless is a serverless inference platform that gives you one OpenAI-compatible API key to call tens of thousands of open-weight models, so you don't provision or manage GPUs yourself.
More detail
- It has three plans: a flat-rate chat plan with unlimited tokens, a per-token developer credit plan for agents and production apps, and custom dedicated-GPU contracts for teams that need guarantees.
- Best for
- Developers running coding agents, app traffic or background automation on open models, and roleplay or creative-writing users who type their own prompts in frontends like SillyTavern, RisuAI or Wyvern.
More detail
- Teams that want long-tail open models through one API without hosting infrastructure.
- Pricing
- Chat is $25/month flat with unlimited tokens, 32K context and 4 concurrent units
More detail
- it is for human-typed interactive use only and bans reselling, app/API traffic, background automation and benchmarking, with misuse cancellable without refund.
- Developer starts at $50 credits per month, billed per toke
- Main limitation
- Resellers, app or API traffic, background automation and benchmarking on the $25 Chat plan, which is restricted to human-driven use.
More detail
- Buyers who need published per-token rates up front, guaranteed capacity on saturated popular models, or a plan they can switch mid-cycle without losing paid time.
- Record
- Living and monitored
- Reviewed
- Evidence
- Public web research
- Machine access
- Structured profile ↗
About
- Featherless is a serverless inference platform that gives you one OpenAI-compatible API key to call tens of thousands of open-weight models, so you don't provision or manage GPUs yourself.
- It has three plans: a flat-rate chat plan with unlimited tokens, a per-token developer credit plan for agents and production apps, and custom dedicated-GPU contracts for teams that need guarantees.
What it helps with
- Running open-weight models normally means provisioning GPUs and managing hosting yourself
- Using many open models usually requires separate per-provider integrations
- Inference spend is hard to predict when you pay per token across many model sizes
- Popular models saturate and fail with no billing consequence, interrupting agent runs
- Long-tail and niche models (regional-language, roleplay-tuned, small) are hard to access through one provider
Capabilities
A practical view of what is available and where limitations apply.
| Offering | Available | Detail |
|---|---|---|
| Open-model catalogue | Yes | Homepage says 40,000+ models; the About page says 30,000+ served, so first-party figures conflict |
| OpenAI-compatible API | Yes | Point OpenAI SDKs or any compatible client at https://api.featherless.ai/v1 with your key |
| Unlimited tokens on Chat plan | Limited | Unlimited length on the $25/month plan, but capped at 32K context and 4 concurrent units, and human-driven use only |
| Per-token developer billing | Yes | From $50 credits/month, input and output priced separately per million tokens; only successful requests are charged |
| Credit rollover | Yes | Unused credits never expire and one-time top-ups are allowed |
| Prompt and completion logging | No | First-party documentation and pricing state prompts are never stored |
| Dedicated GPUs and SLA | Yes | Business tier offers H100, MI325X, B200 and B300 with named engineers and SLA on custom annual contracts |
| Private fine-tunes | Yes | Available on the Business tier only; developer plan cannot host private fine-tunes |
| Guaranteed capacity on popular models | No | A saturated model returns HTTP 503 (not billed); retry or switch to a sibling model |
| Published per-token rates | Unknown | Rates are not shown on the pricing page and must be checked on the models page |
Bot trust
How confidently AI systems can verify this profile from public evidence.
—of 100
ChatGPT—/100
Claude—/100
This profile has not completed a Bot Score assessment yet.
Evidence
Where this record came from, when it was reviewed, and how confidently it can be used.
- Verification
- Scribble researched these facts from the linked public sources.
- The owner has not manually confirmed every field.
- Last reviewed
- Primary source
- featherless.ai
Technical profile details
Record ID
9c741a2b-c376-40b1-95c0-3011471ed48dCanonical URLhttps://profiles.scribble.network/listing/featherless-9c741a2b