Instant, unlimited hosting for any Llama model on HuggingFace.
No servers needed.
Over 2300+ compatible models to choose from.
Starts from just $10/month.
Most Popular Models
The most popular models on the platform in the last 2 weeks.
Trending this Week
The currently trending models that are growing in popularity this week.
Latest Models
These are the most recently available models in Featherless
Trusted by developers at
Simple Pricing Unlimited Tokens
Feather Basic
Max. 15B
$ 10 USD / Month
- Unlimited Personal Use*
- Ever growing list of community models
- New models added weekly
- Access to all models in the Feather ecosystem every month
- Private, secure, and anonymous usage - no logs
Feather Premium
Max. 72B
$ 25 USD / Month
- All the benefits of Feather Basic
- Up to 72B models*
Feather Scale
Max. 72B
$ 75 USD / Month
- All the benefits of Feather Premium
- Scalable pricing to meet your needs
- Host your own private models from Hugging Face
How many concurrencies do you need?
Feather Enterprise
Custom
- Run your entire catalog.
- From your cloud.
- With reduced GPUs.
Frequently Asked Questions
What is Featherless?
Featherless is an LLM hosting provider that offers our subscribers access to a continually expanding library of Hugging Face models.
Featherless: Less hassle, less effort. Start now.
What does it cost?
We offer two pricing plans at $10 and $25 a month.
If the concurrency limits are too restrictive for genuine personal use, please reach out to us via our Discord.
Are my logs stored?
No. As a privacy-focused service, we do not log any of your chats, prompts, or completions. Please see our privacy policy for more details.
Which model architectures are supported?
At present, we support LLaMA-3-based models, including LLaMA-3 and QWEN-2.
Note that QWEN-2 models are only supported up to 16,000 context length.
We plan to add more architectures to our supported list soon.
How do I get new models added?
Ping us on our Discord.
We continuously onboard new models as they become available on Hugging Face.
As we grow, we aim to automate this process to encompass all publicly available Hugging Face models with compatible architectures.
How unlimited is unlimited?
As long as you remain subscribed, there's no time cap on model usage.
To ensure fair individual account use, concurrent requests are limited according to the plan you've selected.
Output is delivered at a speed of 10-40 tokens per second, depending on the model and prompt size.
How can I get in touch?
Join our Discord or find us on r/SillyTavernAI.
Are you running quantized models?
Yes, we use FP8 quantization.
After consulting with the community, we've found that this approach maintains output quality while significantly improving inference speeds.
How are you able to run so many models?
At the heart of the platform is our custom inference stack, in which we can dynamically swap out models on the fly in <1 second for a 10B model.
This allows us to rapidly reconfigure our infrastructure according to user workload and autoscale accordingly, as a single unified unit.
Why would I not just use Hugging Face or etc. directly?
Cost, speed, and customization
While Hugging Face and RunPod let you run any model, they charge $1 per hour or higher for the GPUs. If you plan on using models for over five hours consistently, using our platform is likely the more affordable option.
On the flip side, other providers may offer a limited list of models to optimize for cost and speed. But they may not have the model you want.
Do you have a referral program?
Yes! Refer a friend, and when they subscribe and add your email, both of you get $10 OFF your next monthly bill!
Refer 12 of your friends and you can have a full year off our basic plan! (The discount stacks!)
Details here.
Some of your models don't have open licenses (e.g. CC-BY-NC); how can these be listed on your site?
We are a serverless AI hosting provider. Our product simplifies the process of deploying models from Hugging Face. We are not charging for the underlying models.
We list all supported models, any of which can be made available for inference in milliseconds. We interpret all API requests as model allocation requests and "deploy" the underlying model automatically. This is analogous to how an individual would use RunPod, but at a different scale.
Moreover, we are in contact with model creators of the most popular models to avoid misunderstandings about the nature of Featherless, and have obtained their permission. If you are a model creator and take issue with a listing here, please contact us at [email protected].
Support Open Source AI!
The Recursal AI team's primary goal is to make Open Source AI accessible to everyone, regardless of language or economic status.
That's why we work actively to build open source AI models with the RWKV community that are multi-lingual, highly scalable, and affordable.
The revenue we make from Featherless AI helps support our work in training and scaling such models. You can see our latest open source AI model here.
Also, we encourage you to reach out to your favorite model creators to support them directly as well.