Hugging Face's logo

stepfun-ai
/
Step1X-3D

Diffusers
Safetensors
Model card Files Community
5
ReubenSun commited on
Commit
83e80f8
·
1 Parent(s): cafad5e

init

Browse files
Files changed (21) hide show
  1. Step-1X-3D-Texture/step1x-3d-ig2v.safetensors +3 -0
  2. Step1X-3D-Geometry-1300m/model_index.json +32 -0
  3. Step1X-3D-Geometry-1300m/scheduler/scheduler_config.json +16 -0
  4. Step1X-3D-Geometry-1300m/transformer/config.json +18 -0
  5. Step1X-3D-Geometry-1300m/transformer/diffusion_pytorch_model.safetensors +3 -0
  6. Step1X-3D-Geometry-1300m/vae/config.json +31 -0
  7. Step1X-3D-Geometry-1300m/vae/diffusion_pytorch_model.safetensors +3 -0
  8. Step1X-3D-Geometry-1300m/visual_eature_extractor/preprocessor_config.json +27 -0
  9. Step1X-3D-Geometry-1300m/visual_encoder/config.json +15 -0
  10. Step1X-3D-Geometry-1300m/visual_encoder/diffusion_pytorch_model.safetensors +3 -0
  11. Step1X-3D-Geometry-Label-1300m/label_encoder/config.json +6 -0
  12. Step1X-3D-Geometry-Label-1300m/label_encoder/diffusion_pytorch_model.safetensors +3 -0
  13. Step1X-3D-Geometry-Label-1300m/model_index.json +32 -0
  14. Step1X-3D-Geometry-Label-1300m/scheduler/scheduler_config.json +16 -0
  15. Step1X-3D-Geometry-Label-1300m/transformer/config.json +17 -0
  16. Step1X-3D-Geometry-Label-1300m/transformer/diffusion_pytorch_model.safetensors +3 -0
  17. Step1X-3D-Geometry-Label-1300m/vae/config.json +30 -0
  18. Step1X-3D-Geometry-Label-1300m/vae/diffusion_pytorch_model.safetensors +3 -0
  19. Step1X-3D-Geometry-Label-1300m/visual_eature_extractor/preprocessor_config.json +27 -0
  20. Step1X-3D-Geometry-Label-1300m/visual_encoder/config.json +17 -0
  21. Step1X-3D-Geometry-Label-1300m/visual_encoder/diffusion_pytorch_model.safetensors +3 -0
Step-1X-3D-Texture/step1x-3d-ig2v.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05cd28096e875712c1f2489bb325a9bc7324a41cd3f367d7a38c63f1bea77964
3
+ size 3602537816
Step1X-3D-Geometry-1300m/model_index.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "Step1X3DGeometryPipeline",
3
+ "_diffusers_version": "0.32.2",
4
+ "caption_encoder": [
5
+ null,
6
+ null
7
+ ],
8
+ "label_encoder": [
9
+ null,
10
+ null
11
+ ],
12
+ "scheduler": [
13
+ "diffusers",
14
+ "FlowMatchEulerDiscreteScheduler"
15
+ ],
16
+ "transformer": [
17
+ "step1x3d_geometry.models.transformers.flux_transformer_1d",
18
+ "FluxDenoiser"
19
+ ],
20
+ "vae": [
21
+ "step1x3d_geometry.models.autoencoders.michelangelo_autoencoder",
22
+ "MichelangeloAutoencoder"
23
+ ],
24
+ "visual_eature_extractor": [
25
+ "transformers",
26
+ "BitImageProcessor"
27
+ ],
28
+ "visual_encoder": [
29
+ "step1x3d_geometry.models.conditional_encoders.dinov2_encoder",
30
+ "Dinov2Encoder"
31
+ ]
32
+ }
Step1X-3D-Geometry-1300m/scheduler/scheduler_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "FlowMatchEulerDiscreteScheduler",
3
+ "_diffusers_version": "0.32.2",
4
+ "base_image_seq_len": 256,
5
+ "base_shift": 0.5,
6
+ "invert_sigmas": false,
7
+ "max_image_seq_len": 4096,
8
+ "max_shift": 1.15,
9
+ "num_train_timesteps": 1000,
10
+ "shift": 1.0,
11
+ "shift_terminal": null,
12
+ "use_beta_sigmas": false,
13
+ "use_dynamic_shifting": false,
14
+ "use_exponential_sigmas": false,
15
+ "use_karras_sigmas": false
16
+ }
Step1X-3D-Geometry-1300m/transformer/config.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "weights": null,
3
+ "input_channels": 64,
4
+ "width": 1280,
5
+ "layers": 12,
6
+ "num_single_layers": 24,
7
+ "num_heads": 20,
8
+ "condition_dim": 1024,
9
+ "multi_condition_type": "in_context",
10
+ "use_visual_condition": true,
11
+ "visual_condition_dim": 1024,
12
+ "n_views": 1,
13
+ "use_caption_condition": false,
14
+ "caption_condition_dim": 1024,
15
+ "use_label_condition": false,
16
+ "label_condition_dim": 1024,
17
+ "identity_init": false
18
+ }
Step1X-3D-Geometry-1300m/transformer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8cfc27716009acc60035f2489a9b61de81790625b70aa27a682efc99a5137a9
3
+ size 5265243816
Step1X-3D-Geometry-1300m/vae/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "weights": null,
3
+ "n_samples": 16384,
4
+ "use_downsample": true,
5
+ "downsample_ratio": 0.0625,
6
+ "num_latents": 2048,
7
+ "point_feats": 3,
8
+ "embed_point_feats": false,
9
+ "out_dim": 1,
10
+ "embed_dim": 64,
11
+ "embed_type": "fourier",
12
+ "num_freqs": 8,
13
+ "include_pi": false,
14
+ "width": 768,
15
+ "heads": 12,
16
+ "num_encoder_layers": 8,
17
+ "num_decoder_layers": 16,
18
+ "init_scale": 0.009021097956087902,
19
+ "qkv_bias": false,
20
+ "qk_norm": false,
21
+ "use_ln_post": true,
22
+ "use_flash": true,
23
+ "use_checkpoint": true,
24
+ "use_multi_reso": false,
25
+ "resolutions": null,
26
+ "sampling_prob": null,
27
+ "with_sharp_data": true,
28
+ "volume_decoder_type": "hierarchical",
29
+ "surface_extractor_type": "mc",
30
+ "z_scale_factor": 1.0
31
+ }
Step1X-3D-Geometry-1300m/vae/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa1b17360f6614a54ab3b971c966cfe64ff87a5445186b02c666e634b59c8302
3
+ size 766401652
Step1X-3D-Geometry-1300m/visual_eature_extractor/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 518,
4
+ "width": 518
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "BitImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 518
26
+ }
27
+ }
Step1X-3D-Geometry-1300m/visual_encoder/config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "weights": null,
3
+ "encode_camera": false,
4
+ "camera_embeds_type": "sincos",
5
+ "camera_embeds_dim": null,
6
+ "n_views": 1,
7
+ "empty_embeds_ratio": 0.1,
8
+ "normalize_embeds": false,
9
+ "zero_uncond_embeds": true,
10
+ "freeze_modulation_dino": false,
11
+ "enable_gradient_checkpointing": false,
12
+ "image_size": 518,
13
+ "dino_type": "facebook/dinov2-with-registers-large",
14
+ "kwargs": null
15
+ }
Step1X-3D-Geometry-1300m/visual_encoder/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c07f8abe3af5db214e3341c9f23fb2351cb15591afdae0da8e835dfabc42e850
3
+ size 1217544216
Step1X-3D-Geometry-Label-1300m/label_encoder/config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "hidden_size": 1024,
3
+ "empty_embeds_ratio": 0.1,
4
+ "normalize_embeds": false,
5
+ "zero_uncond_embeds": true
6
+ }
Step1X-3D-Geometry-Label-1300m/label_encoder/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d7d76cfe4bc95f264421fa34076826bada5a218e63febabe69461d8c2607e4b
3
+ size 33104
Step1X-3D-Geometry-Label-1300m/model_index.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "Step1X3DGeometryPipeline",
3
+ "_diffusers_version": "0.32.2",
4
+ "caption_encoder": [
5
+ null,
6
+ null
7
+ ],
8
+ "label_encoder": [
9
+ "step1x3d_geometry.models.conditional_encoders.label_encoder",
10
+ "LabelEncoder"
11
+ ],
12
+ "scheduler": [
13
+ "diffusers",
14
+ "FlowMatchEulerDiscreteScheduler"
15
+ ],
16
+ "transformer": [
17
+ "step1x3d_geometry.models.transformers.flux_transformer_1d",
18
+ "FluxDenoiser"
19
+ ],
20
+ "vae": [
21
+ "step1x3d_geometry.models.autoencoders.michelangelo_autoencoder",
22
+ "MichelangeloAutoencoder"
23
+ ],
24
+ "visual_eature_extractor": [
25
+ "transformers",
26
+ "BitImageProcessor"
27
+ ],
28
+ "visual_encoder": [
29
+ "step1x3d_geometry.models.conditional_encoders.dinov2_clip_encoder",
30
+ "Dinov2CLIPEncoder"
31
+ ]
32
+ }
Step1X-3D-Geometry-Label-1300m/scheduler/scheduler_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "FlowMatchEulerDiscreteScheduler",
3
+ "_diffusers_version": "0.32.2",
4
+ "base_image_seq_len": 256,
5
+ "base_shift": 0.5,
6
+ "invert_sigmas": false,
7
+ "max_image_seq_len": 4096,
8
+ "max_shift": 1.15,
9
+ "num_train_timesteps": 1000,
10
+ "shift": 1.0,
11
+ "shift_terminal": null,
12
+ "use_beta_sigmas": false,
13
+ "use_dynamic_shifting": false,
14
+ "use_exponential_sigmas": false,
15
+ "use_karras_sigmas": false
16
+ }
Step1X-3D-Geometry-Label-1300m/transformer/config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "input_channels": 64,
3
+ "width": 1536,
4
+ "layers": 8,
5
+ "num_single_layers": 16,
6
+ "num_heads": 16,
7
+ "condition_dim": 1024,
8
+ "multi_condition_type": "in_context",
9
+ "use_visual_condition": true,
10
+ "visual_condition_dim": 1024,
11
+ "n_views": 1,
12
+ "use_caption_condition": false,
13
+ "caption_condition_dim": 1024,
14
+ "use_label_condition": true,
15
+ "label_condition_dim": 1024,
16
+ "identity_init": false
17
+ }
Step1X-3D-Geometry-Label-1300m/transformer/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0df096288471320f3fd566e94db628c3a36a697ae6fa55e60152b236fd5918aa
3
+ size 5086980952
Step1X-3D-Geometry-Label-1300m/vae/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "n_samples": 16384,
3
+ "use_downsample": true,
4
+ "downsample_ratio": 0.0625,
5
+ "num_latents": 2048,
6
+ "point_feats": 3,
7
+ "embed_point_feats": false,
8
+ "out_dim": 1,
9
+ "embed_dim": 64,
10
+ "embed_type": "fourier",
11
+ "num_freqs": 8,
12
+ "include_pi": false,
13
+ "width": 768,
14
+ "heads": 12,
15
+ "num_encoder_layers": 8,
16
+ "num_decoder_layers": 16,
17
+ "init_scale": 0.009021097956087902,
18
+ "qkv_bias": false,
19
+ "qk_norm": false,
20
+ "use_ln_post": true,
21
+ "use_flash": true,
22
+ "use_checkpoint": true,
23
+ "use_multi_reso": false,
24
+ "resolutions": null,
25
+ "sampling_prob": null,
26
+ "with_sharp_data": true,
27
+ "volume_decoder_type": "hierarchical",
28
+ "surface_extractor_type": "mc",
29
+ "z_scale_factor": 1.0
30
+ }
Step1X-3D-Geometry-Label-1300m/vae/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9d764b4e3fdf377f1ba960215f2e91800fa6bf86ec5243340b96e22a0a7f781
3
+ size 766401652
Step1X-3D-Geometry-Label-1300m/visual_eature_extractor/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 518,
4
+ "width": 518
5
+ },
6
+ "do_center_crop": true,
7
+ "do_convert_rgb": true,
8
+ "do_normalize": true,
9
+ "do_rescale": true,
10
+ "do_resize": true,
11
+ "image_mean": [
12
+ 0.48145466,
13
+ 0.4578275,
14
+ 0.40821073
15
+ ],
16
+ "image_processor_type": "BitImageProcessor",
17
+ "image_std": [
18
+ 0.26862954,
19
+ 0.26130258,
20
+ 0.27577711
21
+ ],
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 518
26
+ }
27
+ }
Step1X-3D-Geometry-Label-1300m/visual_encoder/config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "encode_camera": false,
3
+ "camera_embeds_type": "sincos",
4
+ "camera_embeds_dim": null,
5
+ "n_views": 1,
6
+ "empty_embeds_ratio": 0.1,
7
+ "normalize_embeds": false,
8
+ "zero_uncond_embeds": true,
9
+ "freeze_modulation_clip": false,
10
+ "freeze_modulation_dino": false,
11
+ "enable_gradient_checkpointing": false,
12
+ "image_size": 518,
13
+ "fuse_type": "concat",
14
+ "dino_type": "facebook/dinov2-with-registers-large",
15
+ "clip_type": "openai/clip-vit-large-patch14",
16
+ "kwargs": null
17
+ }
Step1X-3D-Geometry-Label-1300m/visual_encoder/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9b22f50296f1f63841c92a724c732bffee8051b59e830e9eac9a99af8d0e65e
3
+ size 2928089212