mean_flow_large_2(pretrained: bool = False, overrides: object = {})Large width, patch 2.
Model Size
Parameters
pretrainedbool= FalseAccepted for signature parity with the rest of the zoo. No
MeanFlow checkpoint ships — the paper releases none, and a
True here would have nothing to load.**overridesobject= {}Optional
MeanFlowConfig field overrides.Returns
MeanFlowModelThe average-velocity network.
Notes
Reference: Geng et al., arXiv:2505.13447, 2025, Tables 2 and 4. Reports 3.84 FID at one function evaluation.
The guidance defaults change here: the paper drops the mixing weight
to 0.2 and stops applying guidance above t = 0.8, which is a
tuning of the target rather than of the network.
Examples
>>> from lucid.models import mean_flow_large_2
>>> config = mean_flow_large_2().config
>>> config.depth, config.guidance_interval
(24, (0.0, 0.8))