bridge.recipes.flux.h100.flux#

Module Contents#

Functions#

flux_12b_pretrain_2gpu_h100_bf16_config

Return a pre-training configuration for FLUX 12B model.

flux_12b_sft_2gpu_h100_bf16_config

Return an SFT (supervised fine-tuning) configuration for FLUX 12B model.

Data#

__all__

API#

bridge.recipes.flux.h100.flux.flux_12b_pretrain_2gpu_h100_bf16_config() → megatron.bridge.training.config.ConfigContainer#

Return a pre-training configuration for FLUX 12B model.

Default parallelism: TP=2, PP=1. Uses mock/synthetic data when data_paths is None. To customize (e.g. data paths, checkpoint dir), edit this recipe or add a new recipe that builds on these defaults.

bridge.recipes.flux.h100.flux.flux_12b_sft_2gpu_h100_bf16_config() → megatron.bridge.training.config.ConfigContainer#

Return an SFT (supervised fine-tuning) configuration for FLUX 12B model.

Uses the same defaults as flux_12b_pretrain_config().

bridge.recipes.flux.h100.flux.__all__#

[‘flux_12b_pretrain_2gpu_h100_bf16_config’, ‘flux_12b_sft_2gpu_h100_bf16_config’]