| { |
| "sources": [ |
| { |
| "repository": "hi3d-official/Hi3Dv1.0", |
| "revision": "705fc0901c5d896cdcbf1dbf1feaa173e62bdd3a", |
| "files": { |
| "transformer_512/config.json": "809189b8183c3f44118751322f2cb6fe6fa4ca1922ea16c88cd032970d4bc937", |
| "transformer_512/model.safetensors": "8ba88311dccf23463c6e10dd0d3bd5e0c5c35d1608c6f466413b6428b5ad8e56", |
| "transformer_1024/config.json": "2f96d851844e159523cb6cf5e5b29ddbc1354e222a333232612c19e1eaf278c2", |
| "transformer_1024/model.safetensors": "a2ca5e44dde51cacecd59b42fceeb5f14e75d55e6dd047337752c4e6f281008c", |
| "vae/config.json": "71c629eb188485786227277c7bb5b0fa0a1bfcf07a3d21316174da6b4d04f92f", |
| "vae/model.safetensors": "605d207c77f8db0b33c567a5861252f0090cf3c3604dfda1cfee09e8f0b53f3e", |
| "scheduler/scheduler_config.json": "40cdcdaaa6149dea32b86a5ff730e581b069ae619d7f580444ef503737fb0bf3" |
| } |
| }, |
| { |
| "repository": "facebook/dinov2-large", |
| "revision": "47b73eefe95e8d44ec3623f8890bd894b6ea2d6c", |
| "files": { |
| "config.json": "12df51c069a2dc1305e34ba71ef58bc2407ea553b75f4722a1715c1bce3bbed0", |
| "model.safetensors": "399fba97a95f22c36834418bc69373364a99af3a1153da1c0fb31db567c92e23" |
| } |
| }, |
| { |
| "repository": "openai/clip-vit-large-patch14", |
| "revision": "32bd64288804d66eefd0ccbe215aa642df71cc41", |
| "files": { |
| "config.json": "8a09b467700c58138c29d53c605b34ebc69beaadd13274a8a2af8ad2c2f4032a", |
| "model.safetensors": "a2bf730a0c7debf160f7a6b50b3aaf3703e7e88ac73de7a314903141db026dcb" |
| } |
| } |
| ], |
| "tensors": 1899, |
| "bytes": 10158060544, |
| "dtypes": { |
| "transformer_512": { |
| "float32": 138, |
| "float16": 374 |
| }, |
| "transformer_1024": { |
| "float32": 138, |
| "float16": 374 |
| }, |
| "vae": { |
| "float32": 141 |
| }, |
| "dino": { |
| "float32": 343 |
| }, |
| "clip": { |
| "float32": 391 |
| } |
| }, |
| "skipped": [ |
| "vae.decoder.dec.dec3.up.upsample.inv_grid_coord", |
| "vae.decoder.dec.dec2.up.upsample.inv_grid_coord", |
| "vae.decoder.dec.dec1.up.upsample.inv_grid_coord", |
| "clip:text_model.encoder.layers.10.layer_norm1.weight", |
| "clip:text_model.encoder.layers.1.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.3.mlp.fc2.bias", |
| "clip:text_model.embeddings.position_embedding.weight", |
| "clip:text_model.encoder.layers.9.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.1.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.11.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.9.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.8.layer_norm2.weight", |
| "clip:text_model.encoder.layers.3.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.2.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.9.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.2.layer_norm2.weight", |
| "clip:text_model.encoder.layers.10.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.7.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.5.layer_norm2.weight", |
| "clip:text_model.encoder.layers.3.layer_norm1.bias", |
| "clip:text_model.encoder.layers.8.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.2.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.1.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.5.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.6.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.4.layer_norm1.weight", |
| "clip:text_model.encoder.layers.5.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.6.layer_norm1.weight", |
| "clip:text_model.encoder.layers.1.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.6.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.9.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.7.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.6.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.2.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.3.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.6.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.9.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.9.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.2.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.10.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.9.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.11.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.6.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.9.layer_norm1.weight", |
| "clip:text_model.encoder.layers.8.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.8.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.7.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.8.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.0.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.7.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.11.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.7.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.7.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.7.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.3.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.0.layer_norm1.bias", |
| "clip:text_model.encoder.layers.8.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.2.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.7.self_attn.out_proj.bias", |
| "clip:text_model.final_layer_norm.weight", |
| "clip:text_model.encoder.layers.11.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.7.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.7.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.7.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.8.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.4.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.0.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.8.layer_norm1.weight", |
| "clip:text_model.encoder.layers.7.layer_norm2.bias", |
| "clip:text_model.encoder.layers.3.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.7.layer_norm1.bias", |
| "clip:text_model.encoder.layers.6.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.6.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.0.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.8.layer_norm1.bias", |
| "clip:text_model.encoder.layers.7.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.6.self_attn.out_proj.bias", |
| "clip:text_model.final_layer_norm.bias", |
| "clip:text_model.encoder.layers.6.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.7.layer_norm2.weight", |
| "clip:text_model.encoder.layers.3.layer_norm2.weight", |
| "clip:text_model.encoder.layers.6.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.4.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.10.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.6.layer_norm1.bias", |
| "clip:text_model.encoder.layers.0.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.5.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.5.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.5.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.4.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.5.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.8.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.5.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.5.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.5.layer_norm1.bias", |
| "clip:text_model.encoder.layers.4.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.4.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.4.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.6.layer_norm2.weight", |
| "clip:text_model.encoder.layers.10.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.4.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.4.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.8.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.4.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.2.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.10.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.9.layer_norm2.weight", |
| "clip:text_model.encoder.layers.4.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.4.layer_norm2.bias", |
| "clip:text_model.encoder.layers.4.layer_norm1.bias", |
| "clip:text_model.encoder.layers.7.layer_norm1.weight", |
| "clip:text_model.encoder.layers.11.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.3.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.5.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.2.layer_norm2.bias", |
| "clip:text_model.encoder.layers.3.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.6.layer_norm2.bias", |
| "clip:text_model.encoder.layers.0.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.0.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.8.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.11.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.3.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.3.layer_norm1.weight", |
| "clip:text_model.encoder.layers.9.layer_norm2.bias", |
| "clip:text_model.encoder.layers.2.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.6.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.2.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.1.mlp.fc2.bias", |
| "clip:visual_projection.weight", |
| "clip:text_model.encoder.layers.9.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.2.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.2.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.11.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.0.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.2.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.6.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.3.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.11.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.8.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.11.layer_norm2.weight", |
| "clip:text_model.encoder.layers.11.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.5.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.11.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.10.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.3.layer_norm2.bias", |
| "clip:text_projection.weight", |
| "clip:text_model.encoder.layers.9.layer_norm1.bias", |
| "clip:text_model.encoder.layers.0.layer_norm2.bias", |
| "clip:text_model.encoder.layers.8.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.9.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.0.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.10.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.10.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.9.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.0.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.10.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.2.layer_norm1.bias", |
| "clip:text_model.encoder.layers.1.layer_norm2.weight", |
| "clip:text_model.encoder.layers.10.mlp.fc2.bias", |
| "clip:text_model.encoder.layers.11.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.10.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.10.layer_norm2.bias", |
| "clip:text_model.encoder.layers.1.layer_norm1.weight", |
| "clip:text_model.encoder.layers.11.layer_norm1.bias", |
| "clip:text_model.encoder.layers.11.layer_norm1.weight", |
| "clip:text_model.encoder.layers.10.layer_norm1.bias", |
| "clip:text_model.encoder.layers.1.self_attn.k_proj.weight", |
| "clip:text_model.encoder.layers.0.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.1.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.11.layer_norm2.bias", |
| "clip:text_model.encoder.layers.5.self_attn.v_proj.weight", |
| "clip:text_model.encoder.layers.1.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.1.self_attn.v_proj.bias", |
| "clip:text_model.encoder.layers.8.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.5.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.1.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.4.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.4.layer_norm2.weight", |
| "clip:text_model.encoder.layers.1.self_attn.out_proj.bias", |
| "clip:text_model.encoder.layers.3.self_attn.k_proj.bias", |
| "clip:text_model.encoder.layers.2.layer_norm1.weight", |
| "clip:text_model.encoder.layers.0.self_attn.out_proj.weight", |
| "clip:text_model.encoder.layers.10.layer_norm2.weight", |
| "clip:logit_scale", |
| "clip:text_model.encoder.layers.1.layer_norm2.bias", |
| "clip:text_model.encoder.layers.1.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.4.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.0.self_attn.q_proj.bias", |
| "clip:text_model.encoder.layers.9.self_attn.q_proj.weight", |
| "clip:text_model.encoder.layers.2.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.8.layer_norm2.bias", |
| "clip:text_model.encoder.layers.0.layer_norm1.weight", |
| "clip:text_model.encoder.layers.5.layer_norm2.bias", |
| "clip:text_model.encoder.layers.11.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.3.mlp.fc1.bias", |
| "clip:text_model.encoder.layers.0.layer_norm2.weight", |
| "clip:text_model.embeddings.position_ids", |
| "clip:text_model.encoder.layers.5.layer_norm1.weight", |
| "clip:text_model.encoder.layers.10.mlp.fc2.weight", |
| "clip:text_model.encoder.layers.9.self_attn.v_proj.weight", |
| "clip:text_model.embeddings.token_embedding.weight", |
| "clip:text_model.encoder.layers.3.mlp.fc1.weight", |
| "clip:text_model.encoder.layers.1.layer_norm1.bias" |
| ], |
| "transforms": { |
| "conditioner.dino": "Hugging Face Dinov2Model keys to mlx-vlm dinov2 layout; q/k/v merged into qkv; patch conv (O, I, H, W) -> (O, H, W, I)", |
| "conditioner.clip": "vision_model.* only; position_ids dropped; patch conv (O, I, H, W) -> (O, H, W, I)", |
| "transformer_512, transformer_1024, vae": "values and dtypes unchanged; spconv weights (O, kx, ky, kz, I) used as-is" |
| } |
| } |
|
|
