Instructions to use ByteDance/Bernini-Diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ByteDance/Bernini-Diffusers with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ByteDance/Bernini-Diffusers", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
File size: 1,396 Bytes
7336937 b988d6a 7336937 b988d6a 7336937 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 | {
"model_type": "bernini",
"architectures": ["BerniniModel"],
"mllm_attn_implementation": "sdpa",
"partial_pretrain_model": "",
"base_dir": "ByteDance/Bernini-Diffusers",
"vae_subfolder": "vae",
"cotrain": true,
"boundary_ratio": 0.417,
"switch_dit_boundary": 0.875,
"shift": 3.0,
"logit_mean": 0.5,
"logit_std": 1,
"mode_scale": 1.29,
"scratch": true,
"transformers_version": "4.57.3",
"mllm_subfolder": "mllm",
"processor_subfolder": "mllm",
"bernini_ckpt_subfolder": "bernini",
"scratch_mllm": true,
"use_src_id_rotary_emb": true,
"feature_type_from_stage_one": "masked_tgt_embed_with_qwen_txt_vit_tokens",
"num_mask_token": 4096,
"max_sequence_length": 512,
"clip_diff_cfg": {
"model_type": "flow_match",
"z_channels": 3584,
"target_channels": 3584,
"width": 4096,
"diffusion_batch_mul": 16,
"shift": 2.0
},
"connector_cfg": {
"model_type": "MLPConnector",
"out_dim_for_gen": 4096,
"enable_gen_branch": true,
"out_dim_for_vit": 3584,
"enable_vit_branch": true,
"gen_head_type": "zerolinear",
"zero_init_proj_gen_last": true
},
"mask_ratio_infer_cfg": {
"generator_type": "default"
},
"t5_max_sequence_length": 512,
"t5_text_encoder_subfolder": "t5_text_encoder",
"t5_tokenizer_subfolder": "t5_tokenizer",
"t5_combine_type": "concat_with_zero_init",
"target_fps": 16
}
|