Fine-tuning - verified October 18, 2026

Hugging Face fine-tuning: status, methods and prices

Deprecated. AutoTrain no longer maintained. 1 offering recorded: SFT. Training is priced billed at another rate, in Hugging Face's own units. Weight export: Yes.

Recorded as: Hugging Face.

Read from huggingface.co. Compare every fine-tuning API

Status: Deprecated
Offerings: 1
Methods: SFT
Export weights: Yes

Status

Is Hugging Face fine-tuning still available?

Deprecated

AutoTrain no longer maintained.

“This project is no longer maintained. No new features will be added and bugs will not be fixed.”

Docs - huggingface.co

The offering

Prices are in the provider's own unit, as written. Not stated means the page we checked does not say.

SFTDeprecated

AutoTrain Advanced on Spaces

Supervised fine-tuning

Status: Deprecated (unmaintained)
“This project is no longer maintained. No new features will be added and bugs will not be fixed.”
huggingface.co
Tunable base models: Not stated
Training price: AutoTrain Advanced on Spaces training prices in Hugging Face's own units
Model Price Unit Quote
Spaces hardware per-minute hardware rate per minute of Spaces hardware “Charges are accrued per minute according to the type of hardware utilized during training.”huggingface.co
Tuned-model inference: Not stated
Hosting requirement: Not stated
Export weights: Yes
“you are free to download and deploy your models wherever you choose”
huggingface.co
Data limits: Not stated
Notes: Method coverage (SFT, DPO, etc.) not recorded from fetched pages; 'sft' is a placeholder category.

AutoTrain Advanced on Spaces training prices in Hugging Face's own units
Model	Price	Unit	Quote
Spaces hardware	per-minute hardware rate	per minute of Spaces hardware	“Charges are accrued per minute according to the type of hardware utilized during training.”huggingface.co

Research notes

AutoTrain Advanced is no longer maintained; it runs as a Docker Space billed by Spaces hardware time (or locally for free). No documented per-token fine-tuning price and no documented training product on Inference Endpoints were found; the pricing page has no training/fine-tuning line.

Plain language

The methods, explained

SFTSupervised fine-tuning: You show the model example prompts with the answers you want, and it learns to answer like that. The default method almost everywhere.
DPODirect preference optimization (preference tuning): You give pairs of answers to the same prompt, one preferred and one not, and the model learns to lean toward the preferred kind. Good for tone and style.
RFTReinforcement fine-tuning: The model attempts tasks and a grader (your code, or another model) scores each attempt; training pushes it toward higher-scoring answers. Often billed by training hour, plus the grader's tokens.
Vision SFTVision fine-tuning: Supervised fine-tuning where the examples include images, so the model gets better at a visual task.
DistillationDistillation: A large teacher model writes answers that are used to fine-tune a smaller, cheaper student model. You pay for the teacher's calls as well as the training.
LoRALoRA (adapter) fine-tuning: Trains a small set of extra weights, an adapter, on top of the frozen base model. Cheaper and faster than full fine-tuning; the adapter is served alongside the base model.
FullFull-parameter fine-tuning: Updates every weight in the model. Costs more than LoRA and produces a complete copy of the model.
CPTContinued pre-training: Keeps training the model on a large body of raw, unlabeled text from your domain, so it absorbs vocabulary and facts before any instruction tuning.

FAQ

Does Hugging Face offer fine-tuning?

Deprecated or no longer maintained. AutoTrain no longer maintained. The source says: "This project is no longer maintained. No new features will be added and bugs will not be fixed."

How much does Hugging Face fine-tuning cost?

Billed at another rate: Spaces hardware per-minute hardware rate. Prices are in Hugging Face's own units and are not converted.

Can you download a model you fine-tune on Hugging Face?

Yes. The docs say: "you are free to download and deploy your models wherever you choose"

Sources

Last verified 2026-10-18. This is a directory of what each provider's own pages say, not a quality ranking. Prices are copied in the provider's own unit and never converted, so rows in different units are not comparable.

Newsletter

One email per week. New model releases, deprecations and pricing changes, and what they mean for what you build.