I am getting this error only Inference Providers with accurate pricing are available for organizations

Since today, whenever i try to use API with following strucutre

import os
from openai import OpenAI

client = OpenAI(
base_url=“https://huggingface.co/proxy/router.huggingface.co/v1”,
api_key=“key”,
default_headers={
“X-HF-Bill-To”: “org”
}
)

completion = client.chat.completions.create(
model=“deepseek-ai/DeepSeek-V3-0324:together”,
messages=[
{
“role”: “user”,
“content”: “What is the capital of France?”
}
],
)

print(completion.choices[0].message)

i am getting this error.

only Inference Providers with accurate pricing are available for organizations.

1 Like

I’m not sure if it’s related to the error, but there seems to have been a major change.

Hey @HA01AI Are you still seeing this error message?

1 Like

Hi yes sadly i am still receiving same error.

here is my test code which i copied from huggingface

import os
from openai import OpenAI

client = OpenAI(
base_url=“https://huggingface.co/proxy/router.huggingface.co/v1”,
api_key=“KEY”,
default_headers={
“X-HF-Bill-To”: “artglobal”
}
)

completion = client.chat.completions.create(
model=“deepseek-ai/DeepSeek-V3-0324:together”,
messages=[
{
“role”: “user”,
“content”: “What is the capital of France?”
}
],
)

print(completion.choices[0].message)

1 Like

as user mentioned yes i am facing this error after this change.

1 Like

Yes it is related to this, after this change i am getting error. old api’s still work

1 Like

for this request library based example i am getting the same error

import os
import requests

API_URL = “https://huggingface.co/proxy/router.huggingface.co/v1/chat/completions
headers = {
“Authorization”: f"Bearer hf_*****",
“X-HF-Bill-To”: “artglobal”
}

def query(payload):
response = requests.post(API_URL, headers=headers, json=payload)
return response.json()

response = query({
“messages”: [
{
“role”: “user”,
“content”: “What is the capital of France?”
}
],
“model”: “moonshotai/Kimi-K2-Instruct:together”
})

print(response)

also i like to mention that when i remove the “X-HF-Bill-To”: “artglobal” i don’t get error

1 Like

It looks like your organization account isn’t getting billed or the “X-HF-Bill-To” header isn’t being passed. We have some documentation for billing your organization, with some examples, here: Pricing and Billing.

If you’re still experiencing issues after adjusting the header, let me know!

1 Like

The billing is currently being processed through the old API, where we were required to specify the provider’s name, for example, like this: https://huggingface.co/proxy/router.huggingface.co/fireworks-ai/inference/v1. It used to work perfectly fine, but today, after the recent changes, it started experiencing issues."

1 Like

Hi , i am still facing this issue even after adjusting the header

1 Like

Hi everyone,

our team (3 seats) just switched to Inference Providers and we need to centralize all costs on the credit card linked to our organization.

We’re sending the X‑HF‑Bill‑To: <org-name> header, but every call that targets a model served by Together AI (e.g. deepseek-ai/DeepSeek‑V3-0324:together) fails with

403  Only Inference Providers with accurate pricing are available for organizations

If we drop the header, the call succeeds – but then it’s billed to the individual user, which defeats our governance/compliance workflow.

Requirements

  1. Single credit card at org level – no personal payments/expensing.
  2. Freedom to use any provider without forcing each user to store a separate API key.

Today the docs say PAY‑G via org works “only for PRO users and for integrated providersHugging Face

Is there a temporary way to keep centralized billing without managing separate provider keys + credit cards?

1 Like

Hi @okamirvs can you please email [email protected] with the snippet of code you’re using? Thank you!

1 Like

Hi there - this issue should now be fixed, sorry for the inconvenience!
Can you confirm it works for you now?

2 Likes