Skip to main content
The tables below list every model available through the fine-tuning API. Context lengths are the maximum for that model in SFT and DPO modes. Batch sizes refer to packed batches for text formats. See data preparation for details on packing.
Some models can be fine-tuned but cannot be deployed as dedicated endpoints. To verify deployability before training, confirm the base model appears in the supported models list for dedicated model inference (or run tg beta models configs <BASE_MODEL>). If it isn’t listed there, the fine-tune can’t be hosted on a dedicated endpoint.
Fill out this form to request a model that isn’t in the list.

LoRA fine-tuning

Full fine-tuning

Vision-language

LoRA target modules

​
LoRA fine-tuning

​
Full fine-tuning

​
Vision-language models

For the list of models that support vision-language fine-tuning on image and text data, along with the dataset schema and the train_vision parameter, see vision fine-tuning.

​
LoRA target modules

See LoRA vs. full fine-tuning for the default target modules per model. Pass lora_trainable_modules="all-linear" to train every linear layer.