Skip to content
Free LLM Today

Hugging Face Inference Providers free tier limits

Each line below was read on the official page linked next to it, on the date shown. A line older than 30 days says “not verified” instead of the old value.

Hugging Face · Free tier

Free allowance

$0.10of credits/month

Listed as “subject to change”. Past it, usage needs purchased credits.

Monthly credits for every account, spent on models routed through Hugging Face.

  • Card: not stated
  • Works with OpenAI code
Free allowance
$0.10 of credits/month

Listed as “subject to change”. Past it, usage needs purchased credits.

Requests / min
Not published

The allowance is a monthly credit, not a request count.

Requests / day
Not published

The allowance is a monthly credit, not a request count.

Tokens / min
Not published

The allowance is a monthly credit, not a token count.

Tokens / day
Not published

The allowance is a monthly credit, not a token count.

Card to start
Not verified
Trains on your data
Not verified

Depends on the provider the request is routed to.

Commercial use
Not verified
OpenAI SDK
Yes

The docs use the OpenAI client against the router URL.

Image input
Not verified
Where it is offered
Not verified

7 of 12 cells above were read on the provider’s own page in the last 30 days (oldest check: ). Each check mark opens that page. The other 5 say “not verified” instead of repeating a number from someone else’s article.

What the numbers mean in practice

The free allowance is a monthly credit of $0.10, spent at the price of whichever provider Hugging Face routes the request to. It covers a short evaluation, and the amount is marked “subject to change”.

Requests made with your own provider key are billed by that provider and do not consume the credit.

Getting a Hugging Face Inference Providers key

  1. Create a Hugging Face account.
  2. Create a User Access Token in your settings.
  3. Store it in the HF_TOKEN environment variable.

These steps are a summary of the provider’s console and were not re-walked with a new account; the screens may differ. Open the Hugging Face Inference Providers console.

Test the key with one request

Language
base_url
https://router.huggingface.co/v1
model
deepseek-ai/DeepSeek-V3-0324
key from
$HF_TOKEN
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://router.huggingface.co/v1",
    api_key=os.environ["HF_TOKEN"],
)

reply = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V3-0324",
    messages=[{"role": "user", "content": "Reply with the word: ok"}],
)
print(reply.choices[0].message.content)
Endpoint read on · huggingface.co

The key is read from an environment variable on your machine. This site has no field for keys and never asks for one.

Hugging Face Inference Providers against the others

Questions and answers

What are the Hugging Face Inference Providers free tier limits?

Free tier: $0.10 of credits/month. Requests per minute: not published (The allowance is a monthly credit, not a request count). Requests per day: not published (The allowance is a monthly credit, not a request count). Read on the official page on 3 Oct 2026.

Does Hugging Face Inference Providers require a credit card?

not verified on an official page.

Does Hugging Face Inference Providers train on data sent to the free tier?

not verified on an official page.

Can I use the Hugging Face Inference Providers free tier commercially?

not verified on an official page.

Is Hugging Face Inference Providers compatible with the OpenAI SDK?

Yes. Set the base URL to https://router.huggingface.co/v1 and the model to an id such as deepseek-ai/DeepSeek-V3-0324.