LICENSE
README.md
pyproject.toml
src/megatron/bridge/__init__.py
src/megatron/bridge/package_info.py
src/megatron/bridge/data/__init__.py
src/megatron/bridge/data/base.py
src/megatron/bridge/data/batch_utils.py
src/megatron/bridge/data/conversation_processing.py
src/megatron/bridge/data/iterator_utils.py
src/megatron/bridge/data/loaders.py
src/megatron/bridge/data/samplers.py
src/megatron/bridge/data/sft_processing.py
src/megatron/bridge/data/token_utils.py
src/megatron/bridge/data/utils.py
src/megatron/bridge/data/vlm_processing.py
src/megatron/bridge/data/builders/__init__.py
src/megatron/bridge/data/builders/direct_hf_sft.py
src/megatron/bridge/data/builders/energon.py
src/megatron/bridge/data/builders/gpt_sft.py
src/megatron/bridge/data/builders/mock_vlm_sft.py
src/megatron/bridge/data/collators/__init__.py
src/megatron/bridge/data/collators/registry.py
src/megatron/bridge/data/collators/sequence.py
src/megatron/bridge/data/collators/sequence_padding.py
src/megatron/bridge/data/collators/sft.py
src/megatron/bridge/data/collators/visual.py
src/megatron/bridge/data/datasets/__init__.py
src/megatron/bridge/data/datasets/direct_sft.py
src/megatron/bridge/data/datasets/fim_dataset.py
src/megatron/bridge/data/datasets/gpt_sft.py
src/megatron/bridge/data/datasets/utils.py
src/megatron/bridge/data/energon/__init__.py
src/megatron/bridge/data/energon/base_energon_datamodule.py
src/megatron/bridge/data/energon/hf_task_encoder.py
src/megatron/bridge/data/energon/metadata.py
src/megatron/bridge/data/energon/nemotron_omni_task_encoder.py
src/megatron/bridge/data/energon/prepare.py
src/megatron/bridge/data/energon/task_encoder_utils.py
src/megatron/bridge/data/megatron_mimo/__init__.py
src/megatron/bridge/data/megatron_mimo/base_provider.py
src/megatron/bridge/data/megatron_mimo/collate.py
src/megatron/bridge/data/megatron_mimo/dataset.py
src/megatron/bridge/data/megatron_mimo/dp_utils.py
src/megatron/bridge/data/megatron_mimo/hf_provider.py
src/megatron/bridge/data/megatron_mimo/loaders.py
src/megatron/bridge/data/megatron_mimo/mock_provider.py
src/megatron/bridge/data/packing/__init__.py
src/megatron/bridge/data/packing/algorithms.py
src/megatron/bridge/data/packing/config.py
src/megatron/bridge/data/packing/gpt_sft.py
src/megatron/bridge/data/packing/in_batch.py
src/megatron/bridge/data/packing/offline.py
src/megatron/bridge/data/packing/parquet.py
src/megatron/bridge/data/packing/paths.py
src/megatron/bridge/data/sources/__init__.py
src/megatron/bridge/data/sources/hf.py
src/megatron/bridge/data/sources/hf_adapters.py
src/megatron/bridge/diffusion/__init__.py
src/megatron/bridge/diffusion/base/__init__.py
src/megatron/bridge/diffusion/common/__init__.py
src/megatron/bridge/diffusion/common/dllm.py
src/megatron/bridge/diffusion/common/flow_matching/__init__.py
src/megatron/bridge/diffusion/common/flow_matching/flow_matching_pipeline.py
src/megatron/bridge/diffusion/common/flow_matching/adapters/__init__.py
src/megatron/bridge/diffusion/common/flow_matching/adapters/base.py
src/megatron/bridge/diffusion/common/flow_matching/adapters/simple.py
src/megatron/bridge/diffusion/conversion/__init__.py
src/megatron/bridge/diffusion/conversion/flux/__init__.py
src/megatron/bridge/diffusion/conversion/flux/flux_bridge.py
src/megatron/bridge/diffusion/conversion/flux/flux_hf_pretrained.py
src/megatron/bridge/diffusion/conversion/llada15/__init__.py
src/megatron/bridge/diffusion/conversion/llada15/llada15_bridge.py
src/megatron/bridge/diffusion/conversion/nemotron_labs_diffusion/__init__.py
src/megatron/bridge/diffusion/conversion/nemotron_labs_diffusion/nemotron_labs_diffusion_bridge.py
src/megatron/bridge/diffusion/conversion/wan/__init__.py
src/megatron/bridge/diffusion/conversion/wan/wan_bridge.py
src/megatron/bridge/diffusion/conversion/wan/wan_hf_pretrained.py
src/megatron/bridge/diffusion/data/__init__.py
src/megatron/bridge/diffusion/data/common/__init__.py
src/megatron/bridge/diffusion/data/common/diffusion_energon_datamodule.py
src/megatron/bridge/diffusion/data/common/diffusion_sample.py
src/megatron/bridge/diffusion/data/common/diffusion_task_encoder_with_sp.py
src/megatron/bridge/diffusion/data/common/sequence_packing_utils.py
src/megatron/bridge/diffusion/data/flux/__init__.py
src/megatron/bridge/diffusion/data/flux/flux_energon_datamodule.py
src/megatron/bridge/diffusion/data/flux/flux_mock_datamodule.py
src/megatron/bridge/diffusion/data/flux/flux_taskencoder.py
src/megatron/bridge/diffusion/data/wan/__init__.py
src/megatron/bridge/diffusion/data/wan/wan_energon_datamodule.py
src/megatron/bridge/diffusion/data/wan/wan_mock_datamodule.py
src/megatron/bridge/diffusion/data/wan/wan_taskencoder.py
src/megatron/bridge/diffusion/models/__init__.py
src/megatron/bridge/diffusion/models/common/__init__.py
src/megatron/bridge/diffusion/models/common/dgpt_step.py
src/megatron/bridge/diffusion/models/common/dit_attention.py
src/megatron/bridge/diffusion/models/common/dit_embeddings.py
src/megatron/bridge/diffusion/models/common/nemotron_labs_diffusion_attention.py
src/megatron/bridge/diffusion/models/common/normalization.py
src/megatron/bridge/diffusion/models/flux/__init__.py
src/megatron/bridge/diffusion/models/flux/flux_attention.py
src/megatron/bridge/diffusion/models/flux/flux_layer_spec.py
src/megatron/bridge/diffusion/models/flux/flux_model.py
src/megatron/bridge/diffusion/models/flux/flux_provider.py
src/megatron/bridge/diffusion/models/flux/flux_step.py
src/megatron/bridge/diffusion/models/flux/layers.py
src/megatron/bridge/diffusion/models/flux/flow_matching/__init__.py
src/megatron/bridge/diffusion/models/flux/flow_matching/flux_adapter.py
src/megatron/bridge/diffusion/models/flux/flow_matching/flux_inference_pipeline.py
src/megatron/bridge/diffusion/models/llada15/__init__.py
src/megatron/bridge/diffusion/models/llada15/inference_llada15.py
src/megatron/bridge/diffusion/models/llada15/llada15_attention.py
src/megatron/bridge/diffusion/models/llada15/llada15_provider.py
src/megatron/bridge/diffusion/models/nemotron_labs_diffusion/__init__.py
src/megatron/bridge/diffusion/models/nemotron_labs_diffusion/inference_nemotron_labs_diffusion.py
src/megatron/bridge/diffusion/models/nemotron_labs_diffusion/nemotron_labs_diffusion_provider.py
src/megatron/bridge/diffusion/models/wan/__init__.py
src/megatron/bridge/diffusion/models/wan/rope_utils.py
src/megatron/bridge/diffusion/models/wan/utils.py
src/megatron/bridge/diffusion/models/wan/wan_layer_spec.py
src/megatron/bridge/diffusion/models/wan/wan_model.py
src/megatron/bridge/diffusion/models/wan/wan_provider.py
src/megatron/bridge/diffusion/models/wan/wan_step.py
src/megatron/bridge/diffusion/models/wan/flow_matching/__init__.py
src/megatron/bridge/diffusion/models/wan/flow_matching/flow_inference_pipeline.py
src/megatron/bridge/diffusion/models/wan/flow_matching/flow_matching_pipeline_wan.py
src/megatron/bridge/diffusion/models/wan/inference/__init__.py
src/megatron/bridge/diffusion/models/wan/inference/utils.py
src/megatron/bridge/diffusion/recipes/__init__.py
src/megatron/bridge/diffusion/recipes/flux/__init__.py
src/megatron/bridge/diffusion/recipes/flux/flux.py
src/megatron/bridge/diffusion/recipes/nemotron_labs_diffusion/__init__.py
src/megatron/bridge/diffusion/recipes/nemotron_labs_diffusion/ar_to_dlm.py
src/megatron/bridge/diffusion/recipes/nemotron_labs_diffusion/continuous_pretraining.py
src/megatron/bridge/diffusion/recipes/wan/__init__.py
src/megatron/bridge/diffusion/recipes/wan/wan.py
src/megatron/bridge/inference/_tokenizer.py
src/megatron/bridge/inference/text_generation.py
src/megatron/bridge/inference/vlm/_mcore_compat.py
src/megatron/bridge/inference/vlm/base.py
src/megatron/bridge/inference/vlm/qwenvl_inference_wrapper.py
src/megatron/bridge/inference/vlm/vlm_engine.py
src/megatron/bridge/inference/vlm/vlm_inference_controller.py
src/megatron/bridge/models/__init__.py
src/megatron/bridge/models/_deprecation.py
src/megatron/bridge/models/config.py
src/megatron/bridge/models/distillation_provider.py
src/megatron/bridge/models/gpt_full_te_layer_autocast_spec.py
src/megatron/bridge/models/gpt_provider.py
src/megatron/bridge/models/mla_provider.py
src/megatron/bridge/models/model_provider.py
src/megatron/bridge/models/t5_provider.py
src/megatron/bridge/models/transformer_config.py
src/megatron/bridge/models/bailing/__init__.py
src/megatron/bridge/models/bailing/bailing_moe2_bridge.py
src/megatron/bridge/models/bailing/configuration_bailing_moe_v2.py
src/megatron/bridge/models/bailing/modeling_bailing_moe_v2.py
src/megatron/bridge/models/common/__init__.py
src/megatron/bridge/models/common/base.py
src/megatron/bridge/models/common/te_layers.py
src/megatron/bridge/models/common/unimodal.py
src/megatron/bridge/models/conversion/__init__.py
src/megatron/bridge/models/conversion/auto_bridge.py
src/megatron/bridge/models/conversion/mapping_registry.py
src/megatron/bridge/models/conversion/model_bridge.py
src/megatron/bridge/models/conversion/modelopt_utils.py
src/megatron/bridge/models/conversion/param_mapping.py
src/megatron/bridge/models/conversion/peft_bridge.py
src/megatron/bridge/models/conversion/quant_bridge.py
src/megatron/bridge/models/conversion/quant_mapping.py
src/megatron/bridge/models/conversion/quantization_utils.py
src/megatron/bridge/models/conversion/transformers_compat.py
src/megatron/bridge/models/conversion/utils.py
src/megatron/bridge/models/decorators/__init__.py
src/megatron/bridge/models/decorators/dispatch.py
src/megatron/bridge/models/decorators/torchrun.py
src/megatron/bridge/models/deepseek/__init__.py
src/megatron/bridge/models/deepseek/common.py
src/megatron/bridge/models/deepseek/deepseek_v2_bridge.py
src/megatron/bridge/models/deepseek/deepseek_v3_bridge.py
src/megatron/bridge/models/deepseek/deepseek_v4_bridge.py
src/megatron/bridge/models/deepseek/deepseek_v4_step.py
src/megatron/bridge/models/ernie/__init__.py
src/megatron/bridge/models/ernie/ernie_45_bridge.py
src/megatron/bridge/models/ernie_vl/__init__.py
src/megatron/bridge/models/ernie_vl/ernie45_vl_bridge.py
src/megatron/bridge/models/ernie_vl/ernie45_vl_provider.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/__init__.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/ernie_decoder_layer_spec.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/ernie_moe_layer.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/model.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/vision_attention.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/vision_layer_spec.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/vision_model.py
src/megatron/bridge/models/ernie_vl/modeling_ernie45_vl/vision_transformer_config.py
src/megatron/bridge/models/exaone/__init__.py
src/megatron/bridge/models/exaone/exaone4/__init__.py
src/megatron/bridge/models/exaone/exaone4/exaone4_bridge.py
src/megatron/bridge/models/exaone/exaone4/exaone4_provider.py
src/megatron/bridge/models/exaone/exaone45/__init__.py
src/megatron/bridge/models/exaone/exaone45/exaone45_bridge.py
src/megatron/bridge/models/exaone/exaone45/exaone45_provider.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/__init__.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/model.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/rope.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/text_model.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/transformer_block.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/transformer_config.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/utils.py
src/megatron/bridge/models/exaone/exaone45/modelling_exaone45/vision_model.py
src/megatron/bridge/models/exaone/exaone_moe/__init__.py
src/megatron/bridge/models/exaone/exaone_moe/exaone_moe_bridge.py
src/megatron/bridge/models/exaone/exaone_moe/exaone_moe_provider.py
src/megatron/bridge/models/falcon_h1/__init__.py
src/megatron/bridge/models/falcon_h1/falconh1_bridge.py
src/megatron/bridge/models/falcon_h1/falconh1_provider.py
src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_block.py
src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_layer.py
src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_layer_specs.py
src/megatron/bridge/models/falcon_h1/modeling_falconh1/falconh1_model.py
src/megatron/bridge/models/falcon_h1/modeling_falconh1/mamba_hybrid_layer_allocation.py
src/megatron/bridge/models/gemma/__init__.py
src/megatron/bridge/models/gemma/gemma2_bridge.py
src/megatron/bridge/models/gemma/gemma2_provider.py
src/megatron/bridge/models/gemma/gemma3_bridge.py
src/megatron/bridge/models/gemma/gemma3_provider.py
src/megatron/bridge/models/gemma/gemma4_bridge.py
src/megatron/bridge/models/gemma/gemma4_provider.py
src/megatron/bridge/models/gemma/gemma_bridge.py
src/megatron/bridge/models/gemma/gemma_provider.py
src/megatron/bridge/models/gemma/modeling_gemma4.py
src/megatron/bridge/models/gemma/modules.py
src/megatron/bridge/models/gemma_vl/__init__.py
src/megatron/bridge/models/gemma_vl/gemma3_vl_bridge.py
src/megatron/bridge/models/gemma_vl/gemma3_vl_provider.py
src/megatron/bridge/models/gemma_vl/gemma4_vl_bridge.py
src/megatron/bridge/models/gemma_vl/gemma4_vl_provider.py
src/megatron/bridge/models/gemma_vl/modeling_gemma3_vl.py
src/megatron/bridge/models/gemma_vl/modeling_gemma4_vl.py
src/megatron/bridge/models/gemma_vl/data/__init__.py
src/megatron/bridge/models/gemma_vl/data/collate_fn.py
src/megatron/bridge/models/glm/__init__.py
src/megatron/bridge/models/glm/glm45_bridge.py
src/megatron/bridge/models/glm/glm47_flash_bridge.py
src/megatron/bridge/models/glm_moe_dsa/__init__.py
src/megatron/bridge/models/glm_moe_dsa/glm5_bridge.py
src/megatron/bridge/models/glm_moe_dsa/glm5_provider.py
src/megatron/bridge/models/glm_vl/__init__.py
src/megatron/bridge/models/glm_vl/glm_45v_bridge.py
src/megatron/bridge/models/glm_vl/glm_45v_provider.py
src/megatron/bridge/models/glm_vl/modeling_glm_45v.py
src/megatron/bridge/models/glm_vl/data/__init__.py
src/megatron/bridge/models/glm_vl/data/collate_fn.py
src/megatron/bridge/models/gpt/__init__.py
src/megatron/bridge/models/gpt/gpt_builder.py
src/megatron/bridge/models/gpt_oss/__init__.py
src/megatron/bridge/models/gpt_oss/gpt_oss_bridge.py
src/megatron/bridge/models/hf_pretrained/__init__.py
src/megatron/bridge/models/hf_pretrained/base.py
src/megatron/bridge/models/hf_pretrained/causal_lm.py
src/megatron/bridge/models/hf_pretrained/masked_lm.py
src/megatron/bridge/models/hf_pretrained/safe_config_loader.py
src/megatron/bridge/models/hf_pretrained/state.py
src/megatron/bridge/models/hf_pretrained/utils.py
src/megatron/bridge/models/hy_v3/__init__.py
src/megatron/bridge/models/hy_v3/hy_v3_bridge.py
src/megatron/bridge/models/hybrid/__init__.py
src/megatron/bridge/models/hybrid/hybrid_builder.py
src/megatron/bridge/models/hybrid/hybrid_provider.py
src/megatron/bridge/models/kimi/__init__.py
src/megatron/bridge/models/kimi/kimi_bridge.py
src/megatron/bridge/models/kimi_vl/__init__.py
src/megatron/bridge/models/kimi_vl/kimi_k25_vl_bridge.py
src/megatron/bridge/models/kimi_vl/kimi_k25_vl_provider.py
src/megatron/bridge/models/kimi_vl/modeling_kimi_k25_vl.py
src/megatron/bridge/models/kimi_vl/utils.py
src/megatron/bridge/models/kimi_vl/data/__init__.py
src/megatron/bridge/models/kimi_vl/data/collate_fn.py
src/megatron/bridge/models/llama/__init__.py
src/megatron/bridge/models/llama/llama4_utils.py
src/megatron/bridge/models/llama/llama_bridge.py
src/megatron/bridge/models/llama_nemotron/__init__.py
src/megatron/bridge/models/llama_nemotron/llama_nemotron_bridge.py
src/megatron/bridge/models/llama_nemotron/llama_nemotron_provider.py
src/megatron/bridge/models/mamba/__init__.py
src/megatron/bridge/models/mamba/mamba_builder.py
src/megatron/bridge/models/mamba/mamba_provider.py
src/megatron/bridge/models/megatron_mimo/__init__.py
src/megatron/bridge/models/megatron_mimo/build_model.py
src/megatron/bridge/models/megatron_mimo/llava_provider.py
src/megatron/bridge/models/megatron_mimo/megatron_mimo_builder.py
src/megatron/bridge/models/megatron_mimo/megatron_mimo_config.py
src/megatron/bridge/models/megatron_mimo/megatron_mimo_ddp.py
src/megatron/bridge/models/megatron_mimo/megatron_mimo_provider.py
src/megatron/bridge/models/megatron_mimo/conversion/__init__.py
src/megatron/bridge/models/megatron_mimo/conversion/mimo_model_io.py
src/megatron/bridge/models/megatron_mimo/conversion/orchestrator.py
src/megatron/bridge/models/mimo/__init__.py
src/megatron/bridge/models/mimo/mimo_bridge.py
src/megatron/bridge/models/mimo_v2_flash/__init__.py
src/megatron/bridge/models/mimo_v2_flash/mimo_v2_flash_bridge.py
src/megatron/bridge/models/mimo_v2_flash/mimo_v2_flash_provider.py
src/megatron/bridge/models/mimo_v2_flash/modeling_mimo_v2_flash.py
src/megatron/bridge/models/minimax_m2/__init__.py
src/megatron/bridge/models/minimax_m2/minimax_m2_bridge.py
src/megatron/bridge/models/minimax_m2/minimax_m2_provider.py
src/megatron/bridge/models/minimax_m3/__init__.py
src/megatron/bridge/models/minimax_m3/minimax_m3_bridge.py
src/megatron/bridge/models/minimax_m3/modeling_minimax_m3_vl.py
src/megatron/bridge/models/ministral3/__init__.py
src/megatron/bridge/models/ministral3/ministral3_bridge.py
src/megatron/bridge/models/ministral3/ministral3_provider.py
src/megatron/bridge/models/ministral3/modeling_ministral3.py
src/megatron/bridge/models/ministral3/data/__init__.py
src/megatron/bridge/models/ministral3/data/collate_fn.py
src/megatron/bridge/models/mistral/__init__.py
src/megatron/bridge/models/mistral/mistral_bridge.py
src/megatron/bridge/models/mistral/mistral_provider.py
src/megatron/bridge/models/nemotron/__init__.py
src/megatron/bridge/models/nemotron/nemotron_bridge.py
src/megatron/bridge/models/nemotron_omni/__init__.py
src/megatron/bridge/models/nemotron_omni/modeling_nemotron_omni.py
src/megatron/bridge/models/nemotron_omni/nemotron_omni_bridge.py
src/megatron/bridge/models/nemotron_omni/nemotron_omni_provider.py
src/megatron/bridge/models/nemotron_omni/nemotron_omni_sound.py
src/megatron/bridge/models/nemotron_omni/nemotron_omni_utils.py
src/megatron/bridge/models/nemotron_omni/data/__init__.py
src/megatron/bridge/models/nemotron_omni/data/collate_fn.py
src/megatron/bridge/models/nemotron_vl/__init__.py
src/megatron/bridge/models/nemotron_vl/modeling_nemotron_vl.py
src/megatron/bridge/models/nemotron_vl/nemotron_vl_bridge.py
src/megatron/bridge/models/nemotron_vl/nemotron_vl_provider.py
src/megatron/bridge/models/nemotron_vl/nemotron_vl_utils.py
src/megatron/bridge/models/nemotron_vl/data/__init__.py
src/megatron/bridge/models/nemotron_vl/data/collate_fn.py
src/megatron/bridge/models/nemotronh/__init__.py
src/megatron/bridge/models/nemotronh/nemotron_h_bridge.py
src/megatron/bridge/models/olmoe/__init__.py
src/megatron/bridge/models/olmoe/olmoe_bridge.py
src/megatron/bridge/models/olmoe/olmoe_provider.py
src/megatron/bridge/models/qwen/__init__.py
src/megatron/bridge/models/qwen/qwen2_bridge.py
src/megatron/bridge/models/qwen/qwen35_bridge.py
src/megatron/bridge/models/qwen/qwen3_bridge.py
src/megatron/bridge/models/qwen/qwen3_moe_bridge.py
src/megatron/bridge/models/qwen/qwen3_next_bridge.py
src/megatron/bridge/models/qwen3_asr/__init__.py
src/megatron/bridge/models/qwen3_asr/qwen3_asr_bridge.py
src/megatron/bridge/models/qwen3_asr/qwen3_asr_provider.py
src/megatron/bridge/models/qwen3_asr/hf_qwen3_asr/__init__.py
src/megatron/bridge/models/qwen3_asr/hf_qwen3_asr/configuration_qwen3_asr.py
src/megatron/bridge/models/qwen3_asr/hf_qwen3_asr/modeling_qwen3_asr.py
src/megatron/bridge/models/qwen3_asr/hf_qwen3_asr/processing_qwen3_asr.py
src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/__init__.py
src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/model.py
src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/rope.py
src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/thinker_model.py
src/megatron/bridge/models/qwen3_asr/modeling_qwen3_asr/transformer_config.py
src/megatron/bridge/models/qwen_audio/__init__.py
src/megatron/bridge/models/qwen_audio/modeling_qwen2_audio.py
src/megatron/bridge/models/qwen_audio/qwen2_audio_bridge.py
src/megatron/bridge/models/qwen_audio/qwen2_audio_provider.py
src/megatron/bridge/models/qwen_audio/data/__init__.py
src/megatron/bridge/models/qwen_audio/data/collate_fn.py
src/megatron/bridge/models/qwen_omni/__init__.py
src/megatron/bridge/models/qwen_omni/qwen25_omni_bridge.py
src/megatron/bridge/models/qwen_omni/qwen25_omni_provider.py
src/megatron/bridge/models/qwen_omni/qwen3_omni_bridge.py
src/megatron/bridge/models/qwen_omni/qwen3_omni_provider.py
src/megatron/bridge/models/qwen_omni/qwen3_omni_step.py
src/megatron/bridge/models/qwen_omni/data/__init__.py
src/megatron/bridge/models/qwen_omni/data/collate_fn.py
src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/__init__.py
src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/model.py
src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/rope.py
src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/thinker_model.py
src/megatron/bridge/models/qwen_omni/modeling_qwen25_omni/transformer_config.py
src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/__init__.py
src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/model.py
src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/rope.py
src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/thinker_model.py
src/megatron/bridge/models/qwen_omni/modeling_qwen3_omni/transformer_config.py
src/megatron/bridge/models/qwen_vl/__init__.py
src/megatron/bridge/models/qwen_vl/modeling_qwen25_vl.py
src/megatron/bridge/models/qwen_vl/qwen25_vl_bridge.py
src/megatron/bridge/models/qwen_vl/qwen25_vl_provider.py
src/megatron/bridge/models/qwen_vl/qwen35_vl_bridge.py
src/megatron/bridge/models/qwen_vl/qwen35_vl_provider.py
src/megatron/bridge/models/qwen_vl/qwen3_vl_bridge.py
src/megatron/bridge/models/qwen_vl/qwen3_vl_provider.py
src/megatron/bridge/models/qwen_vl/qwen3_vl_step.py
src/megatron/bridge/models/qwen_vl/data/__init__.py
src/megatron/bridge/models/qwen_vl/data/collate_fn.py
src/megatron/bridge/models/qwen_vl/data/energon.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/__init__.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/attention.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/model.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/rope.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/text_model.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/transformer_block.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/transformer_config.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/utils.py
src/megatron/bridge/models/qwen_vl/modelling_qwen3_vl/vision_model.py
src/megatron/bridge/models/sarvam/__init__.py
src/megatron/bridge/models/sarvam/common.py
src/megatron/bridge/models/sarvam/sarvam_mla_bridge.py
src/megatron/bridge/models/sarvam/sarvam_moe_bridge.py
src/megatron/bridge/models/sarvam/sarvam_provider.py
src/megatron/bridge/models/stepfun/__init__.py
src/megatron/bridge/models/stepfun/configuration_step35.py
src/megatron/bridge/models/stepfun/configuration_step37.py
src/megatron/bridge/models/stepfun/step35_bridge.py
src/megatron/bridge/models/stepfun/step35_provider.py
src/megatron/bridge/models/stepfun/step37_bridge.py
src/megatron/bridge/models/stepfun/step37_flickr8k_step.py
src/megatron/bridge/models/stepfun/step37_provider.py
src/megatron/bridge/models/stepfun/data/__init__.py
src/megatron/bridge/models/stepfun/data/flickr8k/__init__.py
src/megatron/bridge/models/stepfun/data/flickr8k/flickr8k_loader.py
src/megatron/bridge/models/stepfun/data/flickr8k/multimodal_utils.py
src/megatron/bridge/models/stepfun/data/flickr8k/pack_transform.py
src/megatron/bridge/models/stepfun/data/flickr8k/packed_dataloader.py
src/megatron/bridge/models/stepfun/data/flickr8k/packing.py
src/megatron/bridge/models/stepfun/data/flickr8k/preprocess.py
src/megatron/bridge/models/stepfun/data/flickr8k/provider.py
src/megatron/bridge/models/stepfun/data/flickr8k/samplers.py
src/megatron/bridge/models/stepfun/data/flickr8k/template.py
src/megatron/bridge/models/stepfun/modelling_step37/__init__.py
src/megatron/bridge/models/stepfun/modelling_step37/attention.py
src/megatron/bridge/models/stepfun/modelling_step37/image_insert_embedding.py
src/megatron/bridge/models/stepfun/modelling_step37/model.py
src/megatron/bridge/models/stepfun/modelling_step37/rope.py
src/megatron/bridge/models/stepfun/modelling_step37/text_model.py
src/megatron/bridge/models/stepfun/modelling_step37/transformer_block.py
src/megatron/bridge/models/stepfun/modelling_step37/transformer_config.py
src/megatron/bridge/models/stepfun/modelling_step37/utils.py
src/megatron/bridge/models/stepfun/modelling_step37/vision_model.py
src/megatron/bridge/peft/__init__.py
src/megatron/bridge/peft/adapter_wrapper.py
src/megatron/bridge/peft/base.py
src/megatron/bridge/peft/canonical_lora.py
src/megatron/bridge/peft/dora.py
src/megatron/bridge/peft/dora_layers.py
src/megatron/bridge/peft/lora.py
src/megatron/bridge/peft/lora_layers.py
src/megatron/bridge/peft/lora_merge.py
src/megatron/bridge/peft/module_matcher.py
src/megatron/bridge/peft/recompute.py
src/megatron/bridge/peft/utils.py
src/megatron/bridge/peft/walk_utils.py
src/megatron/bridge/perf_recipes/__init__.py
src/megatron/bridge/perf_recipes/_common.py
src/megatron/bridge/perf_recipes/environment.py
src/megatron/bridge/perf_recipes/deepseek/__init__.py
src/megatron/bridge/perf_recipes/deepseek/common.py
src/megatron/bridge/perf_recipes/deepseek/b200/__init__.py
src/megatron/bridge/perf_recipes/deepseek/b200/deepseek_v3.py
src/megatron/bridge/perf_recipes/deepseek/b300/__init__.py
src/megatron/bridge/perf_recipes/deepseek/b300/deepseek_v3.py
src/megatron/bridge/perf_recipes/deepseek/gb200/__init__.py
src/megatron/bridge/perf_recipes/deepseek/gb200/deepseek_v3.py
src/megatron/bridge/perf_recipes/deepseek/gb200/deepseek_v4.py
src/megatron/bridge/perf_recipes/deepseek/gb300/__init__.py
src/megatron/bridge/perf_recipes/deepseek/gb300/deepseek_v3.py
src/megatron/bridge/perf_recipes/deepseek/gb300/deepseek_v4.py
src/megatron/bridge/perf_recipes/deepseek/h100/__init__.py
src/megatron/bridge/perf_recipes/deepseek/h100/deepseek_v3.py
src/megatron/bridge/perf_recipes/deepseek/vr200/__init__.py
src/megatron/bridge/perf_recipes/deepseek/vr200/deepseek_v3.py
src/megatron/bridge/perf_recipes/glm_moe_dsa/__init__.py
src/megatron/bridge/perf_recipes/glm_moe_dsa/gb200/__init__.py
src/megatron/bridge/perf_recipes/glm_moe_dsa/gb200/glm5.py
src/megatron/bridge/perf_recipes/glm_moe_dsa/h100/__init__.py
src/megatron/bridge/perf_recipes/glm_moe_dsa/h100/glm5.py
src/megatron/bridge/perf_recipes/gpt_oss/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/common.py
src/megatron/bridge/perf_recipes/gpt_oss/b200/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/b200/gpt_oss.py
src/megatron/bridge/perf_recipes/gpt_oss/b300/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/b300/gpt_oss.py
src/megatron/bridge/perf_recipes/gpt_oss/gb200/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/gb200/gpt_oss.py
src/megatron/bridge/perf_recipes/gpt_oss/gb300/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/gb300/gpt_oss.py
src/megatron/bridge/perf_recipes/gpt_oss/h100/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/h100/gpt_oss.py
src/megatron/bridge/perf_recipes/gpt_oss/vr200/__init__.py
src/megatron/bridge/perf_recipes/gpt_oss/vr200/gpt_oss.py
src/megatron/bridge/perf_recipes/kimi/__init__.py
src/megatron/bridge/perf_recipes/kimi/common.py
src/megatron/bridge/perf_recipes/kimi/b200/__init__.py
src/megatron/bridge/perf_recipes/kimi/b200/kimi_k2.py
src/megatron/bridge/perf_recipes/kimi/b300/__init__.py
src/megatron/bridge/perf_recipes/kimi/b300/kimi_k2.py
src/megatron/bridge/perf_recipes/kimi/gb200/__init__.py
src/megatron/bridge/perf_recipes/kimi/gb200/kimi_k2.py
src/megatron/bridge/perf_recipes/kimi/gb300/__init__.py
src/megatron/bridge/perf_recipes/kimi/gb300/kimi_k2.py
src/megatron/bridge/perf_recipes/kimi/h100/__init__.py
src/megatron/bridge/perf_recipes/kimi/h100/kimi_k2.py
src/megatron/bridge/perf_recipes/kimi/vr200/__init__.py
src/megatron/bridge/perf_recipes/kimi/vr200/kimi_k2.py
src/megatron/bridge/perf_recipes/llama/__init__.py
src/megatron/bridge/perf_recipes/llama/common.py
src/megatron/bridge/perf_recipes/llama/b200/__init__.py
src/megatron/bridge/perf_recipes/llama/b200/llama3.py
src/megatron/bridge/perf_recipes/llama/b200/llama31.py
src/megatron/bridge/perf_recipes/llama/b300/__init__.py
src/megatron/bridge/perf_recipes/llama/b300/llama3.py
src/megatron/bridge/perf_recipes/llama/b300/llama31.py
src/megatron/bridge/perf_recipes/llama/gb200/__init__.py
src/megatron/bridge/perf_recipes/llama/gb200/llama3.py
src/megatron/bridge/perf_recipes/llama/gb200/llama31.py
src/megatron/bridge/perf_recipes/llama/gb300/__init__.py
src/megatron/bridge/perf_recipes/llama/gb300/llama3.py
src/megatron/bridge/perf_recipes/llama/gb300/llama31.py
src/megatron/bridge/perf_recipes/llama/h100/__init__.py
src/megatron/bridge/perf_recipes/llama/h100/llama3.py
src/megatron/bridge/perf_recipes/llama/h100/llama31.py
src/megatron/bridge/perf_recipes/llama/r100/__init__.py
src/megatron/bridge/perf_recipes/llama/r100/llama3.py
src/megatron/bridge/perf_recipes/llama/vr200/__init__.py
src/megatron/bridge/perf_recipes/llama/vr200/llama3.py
src/megatron/bridge/perf_recipes/llama/vr200/llama31.py
src/megatron/bridge/perf_recipes/nemotronh/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/common.py
src/megatron/bridge/perf_recipes/nemotronh/b200/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/b200/nemotronh.py
src/megatron/bridge/perf_recipes/nemotronh/b300/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/b300/nemotronh.py
src/megatron/bridge/perf_recipes/nemotronh/gb200/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/gb200/nemotronh.py
src/megatron/bridge/perf_recipes/nemotronh/gb300/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/gb300/nemotronh.py
src/megatron/bridge/perf_recipes/nemotronh/h100/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/h100/nemotronh.py
src/megatron/bridge/perf_recipes/nemotronh/vr200/__init__.py
src/megatron/bridge/perf_recipes/nemotronh/vr200/nemotronh.py
src/megatron/bridge/perf_recipes/qwen/__init__.py
src/megatron/bridge/perf_recipes/qwen/common.py
src/megatron/bridge/perf_recipes/qwen/b200/__init__.py
src/megatron/bridge/perf_recipes/qwen/b200/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen/b300/__init__.py
src/megatron/bridge/perf_recipes/qwen/b300/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen/gb200/__init__.py
src/megatron/bridge/perf_recipes/qwen/gb200/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen/gb300/__init__.py
src/megatron/bridge/perf_recipes/qwen/gb300/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen/h100/__init__.py
src/megatron/bridge/perf_recipes/qwen/h100/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen/vr200/__init__.py
src/megatron/bridge/perf_recipes/qwen/vr200/qwen3_moe.py
src/megatron/bridge/perf_recipes/qwen_vl/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/common.py
src/megatron/bridge/perf_recipes/qwen_vl/b200/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/b200/qwen35_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/b200/qwen3_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/b300/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/b300/qwen35_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/gb200/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/gb200/qwen35_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/gb200/qwen3_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/gb300/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/gb300/qwen35_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/gb300/qwen3_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/h100/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/h100/qwen35_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/h100/qwen3_vl.py
src/megatron/bridge/perf_recipes/qwen_vl/vr200/__init__.py
src/megatron/bridge/perf_recipes/qwen_vl/vr200/qwen35_vl.py
src/megatron/bridge/perf_recipes/wan/__init__.py
src/megatron/bridge/perf_recipes/wan/common.py
src/megatron/bridge/perf_recipes/wan/gb200/__init__.py
src/megatron/bridge/perf_recipes/wan/gb200/wan.py
src/megatron/bridge/perf_recipes/wan/h100/__init__.py
src/megatron/bridge/perf_recipes/wan/h100/wan.py
src/megatron/bridge/recipes/__init__.py
src/megatron/bridge/recipes/common.py
src/megatron/bridge/recipes/run_plugins.py
src/megatron/bridge/recipes/deepseek/__init__.py
src/megatron/bridge/recipes/deepseek/deepseek_v2.py
src/megatron/bridge/recipes/deepseek/deepseek_v3.py
src/megatron/bridge/recipes/deepseek/deepseek_v4.py
src/megatron/bridge/recipes/deepseek/gb200/__init__.py
src/megatron/bridge/recipes/deepseek/gb200/deepseek_v4.py
src/megatron/bridge/recipes/deepseek/gb300/__init__.py
src/megatron/bridge/recipes/deepseek/gb300/deepseek_v4.py
src/megatron/bridge/recipes/deepseek/h100/__init__.py
src/megatron/bridge/recipes/deepseek/h100/deepseek_v2.py
src/megatron/bridge/recipes/deepseek/h100/deepseek_v3.py
src/megatron/bridge/recipes/deepseek/h100/deepseek_v4.py
src/megatron/bridge/recipes/exaone/__init__.py
src/megatron/bridge/recipes/exaone/exaone4.py
src/megatron/bridge/recipes/exaone/exaone45.py
src/megatron/bridge/recipes/exaone/exaone_moe.py
src/megatron/bridge/recipes/exaone/h100/__init__.py
src/megatron/bridge/recipes/exaone/h100/exaone4.py
src/megatron/bridge/recipes/exaone/h100/exaone45.py
src/megatron/bridge/recipes/exaone/h100/exaone_moe.py
src/megatron/bridge/recipes/flux/__init__.py
src/megatron/bridge/recipes/flux/h100/__init__.py
src/megatron/bridge/recipes/flux/h100/flux.py
src/megatron/bridge/recipes/gemma/__init__.py
src/megatron/bridge/recipes/gemma/gemma2.py
src/megatron/bridge/recipes/gemma/gemma3.py
src/megatron/bridge/recipes/gemma/gemma4.py
src/megatron/bridge/recipes/gemma/h100/__init__.py
src/megatron/bridge/recipes/gemma/h100/gemma2.py
src/megatron/bridge/recipes/gemma/h100/gemma3.py
src/megatron/bridge/recipes/gemma/h100/gemma4.py
src/megatron/bridge/recipes/gemma3_vl/__init__.py
src/megatron/bridge/recipes/gemma3_vl/gemma3_vl.py
src/megatron/bridge/recipes/gemma3_vl/h100/__init__.py
src/megatron/bridge/recipes/gemma3_vl/h100/gemma3_vl.py
src/megatron/bridge/recipes/gemma4_vl/__init__.py
src/megatron/bridge/recipes/gemma4_vl/gemma4_vl.py
src/megatron/bridge/recipes/gemma4_vl/h100/__init__.py
src/megatron/bridge/recipes/gemma4_vl/h100/gemma4_vl.py
src/megatron/bridge/recipes/glm/__init__.py
src/megatron/bridge/recipes/glm/glm45.py
src/megatron/bridge/recipes/glm/gb200/__init__.py
src/megatron/bridge/recipes/glm/gb200/glm5.py
src/megatron/bridge/recipes/glm/h100/__init__.py
src/megatron/bridge/recipes/glm/h100/glm45.py
src/megatron/bridge/recipes/glm/h100/glm5.py
src/megatron/bridge/recipes/glm_vl/__init__.py
src/megatron/bridge/recipes/glm_vl/glm_45v.py
src/megatron/bridge/recipes/glm_vl/h100/__init__.py
src/megatron/bridge/recipes/glm_vl/h100/glm_45v.py
src/megatron/bridge/recipes/gpt/__init__.py
src/megatron/bridge/recipes/gpt/gpt3_175b.py
src/megatron/bridge/recipes/gpt/vanilla_gpt.py
src/megatron/bridge/recipes/gpt/h100/__init__.py
src/megatron/bridge/recipes/gpt/h100/gpt3_175b.py
src/megatron/bridge/recipes/gpt/h100/vanilla_gpt.py
src/megatron/bridge/recipes/gpt_oss/__init__.py
src/megatron/bridge/recipes/gpt_oss/gpt_oss.py
src/megatron/bridge/recipes/gpt_oss/h100/__init__.py
src/megatron/bridge/recipes/gpt_oss/h100/gpt_oss.py
src/megatron/bridge/recipes/kimi/__init__.py
src/megatron/bridge/recipes/kimi/kimi_k2.py
src/megatron/bridge/recipes/kimi/h100/__init__.py
src/megatron/bridge/recipes/kimi/h100/kimi_k2.py
src/megatron/bridge/recipes/kimi_vl/__init__.py
src/megatron/bridge/recipes/kimi_vl/kimi_k25_vl.py
src/megatron/bridge/recipes/kimi_vl/h100/__init__.py
src/megatron/bridge/recipes/kimi_vl/h100/kimi_k25_vl.py
src/megatron/bridge/recipes/llama/__init__.py
src/megatron/bridge/recipes/llama/llama2.py
src/megatron/bridge/recipes/llama/llama3.py
src/megatron/bridge/recipes/llama/h100/__init__.py
src/megatron/bridge/recipes/llama/h100/llama2.py
src/megatron/bridge/recipes/llama/h100/llama3.py
src/megatron/bridge/recipes/minimax/__init__.py
src/megatron/bridge/recipes/minimax/minimax_m3.py
src/megatron/bridge/recipes/minimax/h100/__init__.py
src/megatron/bridge/recipes/minimax/h100/minimax_m3.py
src/megatron/bridge/recipes/ministral3/__init__.py
src/megatron/bridge/recipes/ministral3/ministral3.py
src/megatron/bridge/recipes/ministral3/h100/__init__.py
src/megatron/bridge/recipes/ministral3/h100/ministral3.py
src/megatron/bridge/recipes/moonlight/__init__.py
src/megatron/bridge/recipes/moonlight/moonlight_16b.py
src/megatron/bridge/recipes/moonlight/h100/__init__.py
src/megatron/bridge/recipes/moonlight/h100/moonlight_16b.py
src/megatron/bridge/recipes/nemotron_omni/__init__.py
src/megatron/bridge/recipes/nemotron_omni/nemotron_omni.py
src/megatron/bridge/recipes/nemotron_omni/h100/__init__.py
src/megatron/bridge/recipes/nemotron_omni/h100/nemotron_omni.py
src/megatron/bridge/recipes/nemotron_vl/__init__.py
src/megatron/bridge/recipes/nemotron_vl/nemotron_nano_v2_vl.py
src/megatron/bridge/recipes/nemotron_vl/h100/__init__.py
src/megatron/bridge/recipes/nemotron_vl/h100/nemotron_nano_v2_vl.py
src/megatron/bridge/recipes/nemotronh/__init__.py
src/megatron/bridge/recipes/nemotronh/nemotron_3_nano.py
src/megatron/bridge/recipes/nemotronh/nemotron_3_nano_4b.py
src/megatron/bridge/recipes/nemotronh/nemotron_3_super.py
src/megatron/bridge/recipes/nemotronh/nemotron_3_ultra.py
src/megatron/bridge/recipes/nemotronh/nemotron_nano_v2.py
src/megatron/bridge/recipes/nemotronh/nemotronh.py
src/megatron/bridge/recipes/nemotronh/gb200/__init__.py
src/megatron/bridge/recipes/nemotronh/gb200/nemotron_3_nano.py
src/megatron/bridge/recipes/nemotronh/gb200/nemotron_3_super.py
src/megatron/bridge/recipes/nemotronh/h100/__init__.py
src/megatron/bridge/recipes/nemotronh/h100/nemotron_3_nano.py
src/megatron/bridge/recipes/nemotronh/h100/nemotron_3_nano_4b.py
src/megatron/bridge/recipes/nemotronh/h100/nemotron_3_super.py
src/megatron/bridge/recipes/nemotronh/h100/nemotron_3_ultra.py
src/megatron/bridge/recipes/nemotronh/h100/nemotron_nano_v2.py
src/megatron/bridge/recipes/nemotronh/h100/nemotronh.py
src/megatron/bridge/recipes/olmoe/__init__.py
src/megatron/bridge/recipes/olmoe/olmoe_7b.py
src/megatron/bridge/recipes/olmoe/h100/__init__.py
src/megatron/bridge/recipes/olmoe/h100/olmoe_7b.py
src/megatron/bridge/recipes/qwen/__init__.py
src/megatron/bridge/recipes/qwen/qwen2.py
src/megatron/bridge/recipes/qwen/qwen3.py
src/megatron/bridge/recipes/qwen/qwen35.py
src/megatron/bridge/recipes/qwen/qwen3_moe.py
src/megatron/bridge/recipes/qwen/qwen3_next.py
src/megatron/bridge/recipes/qwen/gb200/__init__.py
src/megatron/bridge/recipes/qwen/gb200/qwen35.py
src/megatron/bridge/recipes/qwen/gb200/qwen3_moe.py
src/megatron/bridge/recipes/qwen/h100/__init__.py
src/megatron/bridge/recipes/qwen/h100/qwen2.py
src/megatron/bridge/recipes/qwen/h100/qwen3.py
src/megatron/bridge/recipes/qwen/h100/qwen3_moe.py
src/megatron/bridge/recipes/qwen/h100/qwen3_next.py
src/megatron/bridge/recipes/qwen2_audio/__init__.py
src/megatron/bridge/recipes/qwen2_audio/qwen2_audio.py
src/megatron/bridge/recipes/qwen2_audio/h100/__init__.py
src/megatron/bridge/recipes/qwen2_audio/h100/qwen2_audio.py
src/megatron/bridge/recipes/qwen_omni/__init__.py
src/megatron/bridge/recipes/qwen_omni/qwen3_omni.py
src/megatron/bridge/recipes/qwen_omni/h100/__init__.py
src/megatron/bridge/recipes/qwen_omni/h100/qwen3_omni.py
src/megatron/bridge/recipes/qwen_vl/__init__.py
src/megatron/bridge/recipes/qwen_vl/qwen25_vl.py
src/megatron/bridge/recipes/qwen_vl/qwen25_vl_dataset.py
src/megatron/bridge/recipes/qwen_vl/qwen35_vl.py
src/megatron/bridge/recipes/qwen_vl/qwen3_vl.py
src/megatron/bridge/recipes/qwen_vl/h100/__init__.py
src/megatron/bridge/recipes/qwen_vl/h100/qwen25_vl.py
src/megatron/bridge/recipes/qwen_vl/h100/qwen35_vl.py
src/megatron/bridge/recipes/qwen_vl/h100/qwen3_vl.py
src/megatron/bridge/recipes/stepfun/__init__.py
src/megatron/bridge/recipes/stepfun/step35.py
src/megatron/bridge/recipes/stepfun/step37.py
src/megatron/bridge/recipes/stepfun/h100/__init__.py
src/megatron/bridge/recipes/stepfun/h100/step35.py
src/megatron/bridge/recipes/stepfun/h100/step37.py
src/megatron/bridge/recipes/utils/__init__.py
src/megatron/bridge/recipes/utils/dataset_utils.py
src/megatron/bridge/recipes/utils/determinism_utils.py
src/megatron/bridge/recipes/utils/environment_utils.py
src/megatron/bridge/recipes/utils/naming.py
src/megatron/bridge/recipes/utils/optimizer_utils.py
src/megatron/bridge/recipes/utils/tokenizer_utils.py
src/megatron/bridge/recipes/wan/__init__.py
src/megatron/bridge/recipes/wan/h100/__init__.py
src/megatron/bridge/recipes/wan/h100/wan.py
src/megatron/bridge/training/__init__.py
src/megatron/bridge/training/audio_lm_step.py
src/megatron/bridge/training/callbacks.py
src/megatron/bridge/training/checkpointing.py
src/megatron/bridge/training/comm_overlap.py
src/megatron/bridge/training/config.py
src/megatron/bridge/training/distill.py
src/megatron/bridge/training/eval.py
src/megatron/bridge/training/eval_context_parallel_rebinding.py
src/megatron/bridge/training/fault_tolerance.py
src/megatron/bridge/training/finetune.py
src/megatron/bridge/training/flex_dispatcher_backend.py
src/megatron/bridge/training/forward_step_func_types.py
src/megatron/bridge/training/gpt_step.py
src/megatron/bridge/training/initialize.py
src/megatron/bridge/training/inprocess_restart.py
src/megatron/bridge/training/llava_step.py
src/megatron/bridge/training/losses.py
src/megatron/bridge/training/megatron_mimo_parallel_utils.py
src/megatron/bridge/training/megatron_mimo_step.py
src/megatron/bridge/training/mixed_precision.py
src/megatron/bridge/training/model_load_save.py
src/megatron/bridge/training/nemotron_omni_step.py
src/megatron/bridge/training/nvrx_straggler.py
src/megatron/bridge/training/optim.py
src/megatron/bridge/training/pretrain.py
src/megatron/bridge/training/pretrain_megatron_mimo.py
src/megatron/bridge/training/profiling.py
src/megatron/bridge/training/setup.py
src/megatron/bridge/training/setup_megatron_mimo.py
src/megatron/bridge/training/state.py
src/megatron/bridge/training/tensor_inspect.py
src/megatron/bridge/training/train.py
src/megatron/bridge/training/train_megatron_mimo.py
src/megatron/bridge/training/vlm_step.py
src/megatron/bridge/training/mlm_compat/__init__.py
src/megatron/bridge/training/mlm_compat/activations.py
src/megatron/bridge/training/mlm_compat/arguments.py
src/megatron/bridge/training/mlm_compat/model.py
src/megatron/bridge/training/post_training/__init__.py
src/megatron/bridge/training/post_training/checkpointing.py
src/megatron/bridge/training/post_training/distillation.py
src/megatron/bridge/training/tokenizers/__init__.py
src/megatron/bridge/training/tokenizers/config.py
src/megatron/bridge/training/tokenizers/tokenizer.py
src/megatron/bridge/training/utils/__init__.py
src/megatron/bridge/training/utils/batch_utils.py
src/megatron/bridge/training/utils/checkpoint_utils.py
src/megatron/bridge/training/utils/comet_utils.py
src/megatron/bridge/training/utils/config_utils.py
src/megatron/bridge/training/utils/flop_utils.py
src/megatron/bridge/training/utils/log_utils.py
src/megatron/bridge/training/utils/mlflow_utils.py
src/megatron/bridge/training/utils/moe_token_drop.py
src/megatron/bridge/training/utils/omegaconf_utils.py
src/megatron/bridge/training/utils/packed_seq_utils.py
src/megatron/bridge/training/utils/padding_utils.py
src/megatron/bridge/training/utils/pg_utils.py
src/megatron/bridge/training/utils/sig_utils.py
src/megatron/bridge/training/utils/theoretical_memory_utils.py
src/megatron/bridge/training/utils/train_utils.py
src/megatron/bridge/training/utils/visual_inputs.py
src/megatron/bridge/training/utils/wandb_utils.py
src/megatron/bridge/utils/__init__.py
src/megatron/bridge/utils/activation_map.py
src/megatron/bridge/utils/common_utils.py
src/megatron/bridge/utils/cuda_graph.py
src/megatron/bridge/utils/decorators.py
src/megatron/bridge/utils/fusions.py
src/megatron/bridge/utils/import_utils.py
src/megatron/bridge/utils/instantiate_utils.py
src/megatron/bridge/utils/safe_pickle.py
src/megatron/bridge/utils/safe_url.py
src/megatron/bridge/utils/slurm_utils.py
src/megatron/bridge/utils/vocab_utils.py
src/megatron/bridge/utils/yaml_utils.py
src/megatron_bridge.egg-info/PKG-INFO
src/megatron_bridge.egg-info/SOURCES.txt
src/megatron_bridge.egg-info/dependency_links.txt
src/megatron_bridge.egg-info/entry_points.txt
src/megatron_bridge.egg-info/requires.txt
src/megatron_bridge.egg-info/top_level.txt