Skip to content

Hugging Face

Pricing page

868 listings across 197 models

Data policies

from the provider's published policies

Headquarters: FrancePrivacy policy

On training
We don't have their wording on this. Their policy: huggingface.co/docs/inference-providers/security, Hugging Face privacy policy
On keeping your data

We do not store the request body or response when routing requests through Hugging Face.

Source: huggingface.co/docs/inference-providers/security

More from Hugging Face on keeping your data (1)

Logs are kept for debugging purposes for up to 30 days, but no user data or tokens are stored.

Possible sources (2)
On where data is processed

The Company and its servers are located in the United States.

Source: Hugging Face privacy policy

More from Hugging Face on where data is processed (1)

The Company and its servers are located in the United States. Personal Information collected by the Services may be stored and processed in the United States or any other country in which the Company or its affiliates, subsidiaries or agents maintain facilities.

Possible sources (2)
Other statements

For more information on how your data is handled, please refer to the Data Security Policies of each provider.

Possible sources (2)
More statements from Hugging Face (1)

The Hugging Face Hub, which Inference Providers is a feature of, is SOC2 Type 2 certified.

Possible sources (2)

We take these statements from Hugging Face's own published policies each time we refresh the data, so they can change or disappear when those pages do. Check with the provider before you rely on any of this.

Page 1 of 8. The lists below split this page's listings into those sold by Hugging Face, run by Hugging Face, or sent to Hugging Face by a marketplace.

Sold by Hugging Face

284 listings on this page sold by Hugging Face.

Default order. Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
Kimi K3.1 Pro
$2.00$10.001M
DeepSeek V4.1 Flash 1017
$0.30$1.201M
Kimi K3
$3.00$15.001M
Qwen3.8 MaxQwen/Qwen3.8-2.4T-A95B
$2.50$6.25262K
GLM 5.3
$1.40$4.401M
DeepSeek V4.1 Flash
$0.30$1.201M
Hy 4
$0.417$1.25051MNot measured
DeepSeek V4 Pro 0813
$1.32$3.961M
Inkling 2 Small
$0.30$1.001MNot measuredNot measured
$0.44$1.321MNot measuredNot measured
Qwen 3.8 Flash NextQwen/Qwen3.8-Flash-Next-FP8 Β· fp8
Inference by Nscale
$0.15$0.471MNot enough data yetNot measuredNot measured
Hy4 preview
$0.834$2.5011MNot measured
Ling 3.1 Flash
Inference by Novita AI
Not enough data yet$0.30$0.901MNot measuredNot measured
MiniMax M3.1 Flash
$0.15$0.601MNot measuredNot measured
MiMo V2.6 FlashXiaomiMiMo/MiMo-V2.6-Flash-RL
Inference by Cohere
$0.14$0.281MNot measured
Inference by Baseten
$1.25$4.251MNot measured
GLM 5.2
$1.40$4.40262K
GLM 5.3 Flash
$0.15$0.501M
Kimi K2.6
$0.95cache read $0.16$4.00262K
DeepSeek V4 Flash 0731
$0.14$0.281M
Hy4 Flash
$0.035$0.14051MNot measured
Qwen3 Embedding 8B
Not enough data yet$0.01β€”33KNot enough data yetNot measuredNot measured
Inkling 2 Mini
Inference by Ionet
$0.08$0.35262KNot enough data yetNot measuredNot measured
GLM 5.1
$1.00cache read $0.20$3.20203K
DeepSeek V4 Pro
$0.435cache read $0.0036$0.871M
Show all 284 listings

Showing 25 of 284: one listing for each model.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

How this data works