v3

OpenAPI 3.1.0MIT2026-07-31192456807.8 KB
Fine-tuning

Get model limits

Get model limits for a specific fine-tuning model.

get/fine-tunes/models/limits

Query parameters

model_namestring required

The model name to get limits for.

Response

Model limits.

model_namestring required

The name of the model.

max_num_epochsinteger required

Maximum number of training epochs.

max_num_evalsinteger required

Maximum number of evaluations.

max_learning_ratenumber required

Maximum learning rate.

min_learning_ratenumber required

Minimum learning rate.

supports_full_trainingboolean required

Whether the model supports full (non-LoRA) fine-tuning. When false, only LoRA fine-tuning is available and the full_training limits are reported as zero.

supports_visionboolean required

Whether the model supports vision/multimodal inputs.

supports_toolsboolean required

Whether the model supports tool/function calling.

supports_reasoningboolean required

Whether the model supports reasoning.

merge_output_loraboolean required

Whether a merged checkpoint (the base model with the trained LoRA adapter fused in) is produced for LoRA fine-tunes of this model, in addition to the standalone adapter.

default_gradient_accumulation_stepsinteger required

Default gradient accumulation steps used when a fine-tune request omits the value or sets it to 0.

max_num_checkpointsinteger required

Maximum number of checkpoints that can be saved during a fine-tuning job.

min_max_seq_lengthinteger required

Minimum value allowed for the max_seq_length hyperparameter.

max_seq_length_sftinteger required

Maximum sequence length supported for SFT training.

max_seq_length_dpointeger required

Maximum sequence length supported for DPO training.