Models

List Models

Get all available VLA model types that can be fine-tuned.

from qualia import Qualia

client = Qualia()

models = client.models.list()

Response

VLAModel[]

[
{
  "id": "smolvla",
  "name": "SmolVLA",
  "description": "Lightweight VLA model for efficient training",
  "base_model_id": "lerobot/smolvla_base",
  "camera_slots": ["cam_1", "cam_2", "cam_3"],
  "supports_custom_model": true
},
{
  "id": "pi0",
  "name": "Pi0",
  "description": "Physical Intelligence Pi0 model",
  "base_model_id": "lerobot/pi0_base",
  "camera_slots": ["cam_1", "cam_2", "cam_3"],
  "supports_custom_model": true
},
{
  "id": "pi05",
  "name": "Pi0.5",
  "description": "Physical Intelligence Pi0.5 model",
  "base_model_id": "lerobot/pi05_base",
  "camera_slots": ["cam_1", "cam_2", "cam_3"],
  "supports_custom_model": true
},
{
  "id": "act",
  "name": "ACT",
  "description": "Action Chunking Transformer model",
  "base_model_id": null,
  "camera_slots": ["cam_1", "cam_2", "cam_3"],
  "supports_custom_model": false
},
{
  "id": "gr00t_n1_5",
  "name": "GR00T N1.5",
  "description": "NVIDIA GR00T N1.5 foundation model",
  "base_model_id": null,
  "camera_slots": ["cam_1", "cam_2", "cam_3"],
  "supports_custom_model": false
},
{
  "id": "sarm",
  "name": "SARM",
  "description": "SARM reward model for Reward-Aware Behavior Cloning",
  "base_model_id": null,
  "camera_slots": ["cam_1"],
  "supports_custom_model": false
}
]

Available Model Types

Type	Description	Custom Model	RA-BC Support
smolvla	Lightweight VLA model	Yes	Yes
pi0	Physical Intelligence Pi0 model	Yes	Yes
pi05	Physical Intelligence Pi0.5 model	Yes	Yes
act	Action Chunking Transformer model	No	No
gr00t_n1_5	NVIDIA GR00T N1.5 foundation model	No	No
sarm	SARM reward model.	No	No

The camera_slots field indicates which camera inputs the model expects. Use these slot names as keys in camera_mappings when creating finetune jobs.

Models with supports_custom_model: false use a fixed base model and do not accept a model_id parameter.