WaveCut's picture
Add files using upload-large-folder tool
ddb67eb verified
Raw
History Blame Contribute Delete
22 kB
{
"effective_config": {
"add_skip_keys": true,
"dequantize_fp32": true,
"dynamic_loss_threshold": 0.01,
"group_size": 0,
"is_integer": true,
"is_training": false,
"modules_dtype_dict": {
"int5": [
"model.visual.blocks.0.attn.qkv.weight",
"model.visual.blocks.0.mlp.linear_fc1.weight",
"model.visual.blocks.0.mlp.linear_fc2.weight",
"model.visual.blocks.1.attn.qkv.weight",
"model.visual.blocks.1.mlp.linear_fc1.weight",
"model.visual.blocks.1.mlp.linear_fc2.weight",
"model.visual.blocks.2.mlp.linear_fc2.weight",
"model.visual.blocks.3.mlp.linear_fc2.weight",
"model.visual.blocks.4.mlp.linear_fc2.weight",
"model.visual.blocks.5.mlp.linear_fc2.weight",
"model.visual.blocks.6.mlp.linear_fc2.weight",
"model.visual.blocks.7.mlp.linear_fc2.weight",
"model.visual.blocks.8.mlp.linear_fc2.weight",
"model.visual.blocks.9.mlp.linear_fc2.weight",
"model.visual.blocks.10.mlp.linear_fc2.weight",
"model.visual.blocks.11.mlp.linear_fc2.weight",
"model.visual.blocks.12.mlp.linear_fc2.weight",
"model.visual.blocks.13.mlp.linear_fc2.weight",
"model.visual.blocks.14.mlp.linear_fc2.weight",
"model.visual.blocks.15.mlp.linear_fc2.weight",
"model.visual.blocks.16.mlp.linear_fc2.weight",
"model.visual.blocks.17.mlp.linear_fc2.weight",
"model.visual.blocks.18.mlp.linear_fc2.weight",
"model.visual.blocks.19.mlp.linear_fc2.weight",
"model.visual.blocks.20.mlp.linear_fc2.weight",
"model.visual.blocks.21.mlp.linear_fc2.weight",
"model.visual.blocks.22.mlp.linear_fc2.weight",
"model.visual.blocks.23.mlp.linear_fc2.weight",
"model.visual.blocks.24.mlp.linear_fc2.weight",
"model.visual.blocks.25.mlp.linear_fc2.weight",
"model.visual.blocks.26.mlp.linear_fc2.weight"
],
"uint4": [
"model.visual.blocks.0.attn.proj.weight",
"model.visual.blocks.1.attn.proj.weight",
"model.visual.blocks.2.attn.qkv.weight",
"model.visual.blocks.2.attn.proj.weight",
"model.visual.blocks.2.mlp.linear_fc1.weight",
"model.visual.blocks.3.attn.qkv.weight",
"model.visual.blocks.3.attn.proj.weight",
"model.visual.blocks.3.mlp.linear_fc1.weight",
"model.visual.blocks.4.attn.qkv.weight",
"model.visual.blocks.4.attn.proj.weight",
"model.visual.blocks.4.mlp.linear_fc1.weight",
"model.visual.blocks.5.attn.qkv.weight",
"model.visual.blocks.5.attn.proj.weight",
"model.visual.blocks.5.mlp.linear_fc1.weight",
"model.visual.blocks.6.attn.qkv.weight",
"model.visual.blocks.6.attn.proj.weight",
"model.visual.blocks.6.mlp.linear_fc1.weight",
"model.visual.blocks.7.attn.qkv.weight",
"model.visual.blocks.7.attn.proj.weight",
"model.visual.blocks.7.mlp.linear_fc1.weight",
"model.visual.blocks.8.attn.qkv.weight",
"model.visual.blocks.8.attn.proj.weight",
"model.visual.blocks.8.mlp.linear_fc1.weight",
"model.visual.blocks.9.attn.qkv.weight",
"model.visual.blocks.9.attn.proj.weight",
"model.visual.blocks.9.mlp.linear_fc1.weight",
"model.visual.blocks.10.attn.qkv.weight",
"model.visual.blocks.10.attn.proj.weight",
"model.visual.blocks.10.mlp.linear_fc1.weight",
"model.visual.blocks.11.attn.qkv.weight",
"model.visual.blocks.11.attn.proj.weight",
"model.visual.blocks.11.mlp.linear_fc1.weight",
"model.visual.blocks.12.attn.qkv.weight",
"model.visual.blocks.12.attn.proj.weight",
"model.visual.blocks.12.mlp.linear_fc1.weight",
"model.visual.blocks.13.attn.qkv.weight",
"model.visual.blocks.13.attn.proj.weight",
"model.visual.blocks.13.mlp.linear_fc1.weight",
"model.visual.blocks.14.attn.qkv.weight",
"model.visual.blocks.14.attn.proj.weight",
"model.visual.blocks.14.mlp.linear_fc1.weight",
"model.visual.blocks.15.attn.qkv.weight",
"model.visual.blocks.15.attn.proj.weight",
"model.visual.blocks.15.mlp.linear_fc1.weight",
"model.visual.blocks.16.attn.qkv.weight",
"model.visual.blocks.16.attn.proj.weight",
"model.visual.blocks.16.mlp.linear_fc1.weight",
"model.visual.blocks.17.attn.qkv.weight",
"model.visual.blocks.17.attn.proj.weight",
"model.visual.blocks.17.mlp.linear_fc1.weight",
"model.visual.blocks.18.attn.qkv.weight",
"model.visual.blocks.18.attn.proj.weight",
"model.visual.blocks.18.mlp.linear_fc1.weight",
"model.visual.blocks.19.attn.qkv.weight",
"model.visual.blocks.19.attn.proj.weight",
"model.visual.blocks.19.mlp.linear_fc1.weight",
"model.visual.blocks.20.attn.qkv.weight",
"model.visual.blocks.20.attn.proj.weight",
"model.visual.blocks.20.mlp.linear_fc1.weight",
"model.visual.blocks.21.attn.qkv.weight",
"model.visual.blocks.21.attn.proj.weight",
"model.visual.blocks.21.mlp.linear_fc1.weight",
"model.visual.blocks.22.attn.qkv.weight",
"model.visual.blocks.22.attn.proj.weight",
"model.visual.blocks.22.mlp.linear_fc1.weight",
"model.visual.blocks.23.attn.qkv.weight",
"model.visual.blocks.23.attn.proj.weight",
"model.visual.blocks.23.mlp.linear_fc1.weight",
"model.visual.blocks.24.attn.qkv.weight",
"model.visual.blocks.24.attn.proj.weight",
"model.visual.blocks.24.mlp.linear_fc1.weight",
"model.visual.blocks.25.attn.qkv.weight",
"model.visual.blocks.25.attn.proj.weight",
"model.visual.blocks.25.mlp.linear_fc1.weight",
"model.visual.blocks.26.attn.qkv.weight",
"model.visual.blocks.26.attn.proj.weight",
"model.visual.blocks.26.mlp.linear_fc1.weight",
"model.visual.merger.linear_fc1.weight",
"model.visual.merger.linear_fc2.weight",
"model.visual.deepstack_merger_list.0.linear_fc1.weight",
"model.visual.deepstack_merger_list.0.linear_fc2.weight",
"model.visual.deepstack_merger_list.1.linear_fc1.weight",
"model.visual.deepstack_merger_list.1.linear_fc2.weight",
"model.visual.deepstack_merger_list.2.linear_fc1.weight",
"model.visual.deepstack_merger_list.2.linear_fc2.weight",
"model.language_model.layers.0.self_attn.q_proj.weight",
"model.language_model.layers.0.self_attn.k_proj.weight",
"model.language_model.layers.0.self_attn.v_proj.weight",
"model.language_model.layers.0.mlp.gate_proj.weight",
"model.language_model.layers.0.mlp.up_proj.weight",
"model.language_model.layers.1.self_attn.q_proj.weight",
"model.language_model.layers.1.self_attn.k_proj.weight",
"model.language_model.layers.1.self_attn.v_proj.weight",
"model.language_model.layers.1.mlp.gate_proj.weight",
"model.language_model.layers.1.mlp.up_proj.weight",
"model.language_model.layers.2.self_attn.q_proj.weight",
"model.language_model.layers.2.self_attn.k_proj.weight",
"model.language_model.layers.2.self_attn.v_proj.weight",
"model.language_model.layers.2.mlp.gate_proj.weight",
"model.language_model.layers.2.mlp.up_proj.weight",
"model.language_model.layers.3.self_attn.q_proj.weight",
"model.language_model.layers.3.self_attn.k_proj.weight",
"model.language_model.layers.3.self_attn.v_proj.weight",
"model.language_model.layers.3.mlp.gate_proj.weight",
"model.language_model.layers.3.mlp.up_proj.weight",
"model.language_model.layers.4.self_attn.q_proj.weight",
"model.language_model.layers.4.self_attn.k_proj.weight",
"model.language_model.layers.4.self_attn.v_proj.weight",
"model.language_model.layers.4.mlp.gate_proj.weight",
"model.language_model.layers.4.mlp.up_proj.weight",
"model.language_model.layers.5.self_attn.q_proj.weight",
"model.language_model.layers.5.self_attn.k_proj.weight",
"model.language_model.layers.5.self_attn.v_proj.weight",
"model.language_model.layers.5.mlp.gate_proj.weight",
"model.language_model.layers.5.mlp.up_proj.weight",
"model.language_model.layers.6.self_attn.q_proj.weight",
"model.language_model.layers.6.self_attn.k_proj.weight",
"model.language_model.layers.6.self_attn.v_proj.weight",
"model.language_model.layers.6.mlp.gate_proj.weight",
"model.language_model.layers.6.mlp.up_proj.weight",
"model.language_model.layers.7.self_attn.q_proj.weight",
"model.language_model.layers.7.self_attn.k_proj.weight",
"model.language_model.layers.7.self_attn.v_proj.weight",
"model.language_model.layers.7.mlp.gate_proj.weight",
"model.language_model.layers.7.mlp.up_proj.weight",
"model.language_model.layers.8.self_attn.q_proj.weight",
"model.language_model.layers.8.self_attn.k_proj.weight",
"model.language_model.layers.8.self_attn.v_proj.weight",
"model.language_model.layers.8.mlp.gate_proj.weight",
"model.language_model.layers.8.mlp.up_proj.weight",
"model.language_model.layers.9.self_attn.q_proj.weight",
"model.language_model.layers.9.self_attn.k_proj.weight",
"model.language_model.layers.9.self_attn.v_proj.weight",
"model.language_model.layers.9.mlp.gate_proj.weight",
"model.language_model.layers.9.mlp.up_proj.weight",
"model.language_model.layers.10.self_attn.q_proj.weight",
"model.language_model.layers.10.self_attn.k_proj.weight",
"model.language_model.layers.10.self_attn.v_proj.weight",
"model.language_model.layers.10.mlp.gate_proj.weight",
"model.language_model.layers.10.mlp.up_proj.weight",
"model.language_model.layers.11.self_attn.q_proj.weight",
"model.language_model.layers.11.self_attn.k_proj.weight",
"model.language_model.layers.11.self_attn.v_proj.weight",
"model.language_model.layers.11.mlp.gate_proj.weight",
"model.language_model.layers.11.mlp.up_proj.weight",
"model.language_model.layers.12.self_attn.q_proj.weight",
"model.language_model.layers.12.self_attn.k_proj.weight",
"model.language_model.layers.12.self_attn.v_proj.weight",
"model.language_model.layers.12.mlp.gate_proj.weight",
"model.language_model.layers.12.mlp.up_proj.weight",
"model.language_model.layers.13.self_attn.q_proj.weight",
"model.language_model.layers.13.self_attn.k_proj.weight",
"model.language_model.layers.13.self_attn.v_proj.weight",
"model.language_model.layers.13.mlp.gate_proj.weight",
"model.language_model.layers.13.mlp.up_proj.weight",
"model.language_model.layers.14.self_attn.q_proj.weight",
"model.language_model.layers.14.self_attn.k_proj.weight",
"model.language_model.layers.14.self_attn.v_proj.weight",
"model.language_model.layers.14.mlp.gate_proj.weight",
"model.language_model.layers.14.mlp.up_proj.weight",
"model.language_model.layers.15.self_attn.q_proj.weight",
"model.language_model.layers.15.self_attn.k_proj.weight",
"model.language_model.layers.15.self_attn.v_proj.weight",
"model.language_model.layers.15.mlp.gate_proj.weight",
"model.language_model.layers.15.mlp.up_proj.weight",
"model.language_model.layers.16.self_attn.q_proj.weight",
"model.language_model.layers.16.self_attn.k_proj.weight",
"model.language_model.layers.16.self_attn.v_proj.weight",
"model.language_model.layers.16.mlp.gate_proj.weight",
"model.language_model.layers.16.mlp.up_proj.weight",
"model.language_model.layers.17.self_attn.q_proj.weight",
"model.language_model.layers.17.self_attn.k_proj.weight",
"model.language_model.layers.17.self_attn.v_proj.weight",
"model.language_model.layers.17.mlp.gate_proj.weight",
"model.language_model.layers.17.mlp.up_proj.weight",
"model.language_model.layers.18.self_attn.q_proj.weight",
"model.language_model.layers.18.self_attn.k_proj.weight",
"model.language_model.layers.18.self_attn.v_proj.weight",
"model.language_model.layers.18.mlp.gate_proj.weight",
"model.language_model.layers.18.mlp.up_proj.weight",
"model.language_model.layers.19.self_attn.q_proj.weight",
"model.language_model.layers.19.self_attn.k_proj.weight",
"model.language_model.layers.19.self_attn.v_proj.weight",
"model.language_model.layers.19.mlp.gate_proj.weight",
"model.language_model.layers.19.mlp.up_proj.weight",
"model.language_model.layers.20.self_attn.q_proj.weight",
"model.language_model.layers.20.self_attn.k_proj.weight",
"model.language_model.layers.20.self_attn.v_proj.weight",
"model.language_model.layers.20.mlp.gate_proj.weight",
"model.language_model.layers.20.mlp.up_proj.weight",
"model.language_model.layers.21.self_attn.q_proj.weight",
"model.language_model.layers.21.self_attn.k_proj.weight",
"model.language_model.layers.21.self_attn.v_proj.weight",
"model.language_model.layers.21.mlp.gate_proj.weight",
"model.language_model.layers.21.mlp.up_proj.weight",
"model.language_model.layers.22.self_attn.q_proj.weight",
"model.language_model.layers.22.self_attn.k_proj.weight",
"model.language_model.layers.22.self_attn.v_proj.weight",
"model.language_model.layers.22.mlp.gate_proj.weight",
"model.language_model.layers.22.mlp.up_proj.weight",
"model.language_model.layers.23.self_attn.q_proj.weight",
"model.language_model.layers.23.self_attn.k_proj.weight",
"model.language_model.layers.23.self_attn.v_proj.weight",
"model.language_model.layers.23.mlp.gate_proj.weight",
"model.language_model.layers.23.mlp.up_proj.weight",
"model.language_model.layers.24.self_attn.q_proj.weight",
"model.language_model.layers.24.self_attn.k_proj.weight",
"model.language_model.layers.24.self_attn.v_proj.weight",
"model.language_model.layers.24.mlp.gate_proj.weight",
"model.language_model.layers.24.mlp.up_proj.weight",
"model.language_model.layers.25.self_attn.q_proj.weight",
"model.language_model.layers.25.self_attn.k_proj.weight",
"model.language_model.layers.25.self_attn.v_proj.weight",
"model.language_model.layers.25.mlp.gate_proj.weight",
"model.language_model.layers.25.mlp.up_proj.weight",
"model.language_model.layers.26.self_attn.q_proj.weight",
"model.language_model.layers.26.self_attn.k_proj.weight",
"model.language_model.layers.26.self_attn.v_proj.weight",
"model.language_model.layers.26.mlp.gate_proj.weight",
"model.language_model.layers.26.mlp.up_proj.weight",
"model.language_model.layers.27.self_attn.q_proj.weight",
"model.language_model.layers.27.self_attn.k_proj.weight",
"model.language_model.layers.27.self_attn.v_proj.weight",
"model.language_model.layers.27.mlp.gate_proj.weight",
"model.language_model.layers.27.mlp.up_proj.weight",
"model.language_model.layers.28.self_attn.q_proj.weight",
"model.language_model.layers.28.self_attn.k_proj.weight",
"model.language_model.layers.28.self_attn.v_proj.weight",
"model.language_model.layers.28.mlp.gate_proj.weight",
"model.language_model.layers.28.mlp.up_proj.weight",
"model.language_model.layers.29.self_attn.q_proj.weight",
"model.language_model.layers.29.self_attn.k_proj.weight",
"model.language_model.layers.29.self_attn.v_proj.weight",
"model.language_model.layers.29.mlp.gate_proj.weight",
"model.language_model.layers.29.mlp.up_proj.weight",
"model.language_model.layers.30.self_attn.q_proj.weight",
"model.language_model.layers.30.self_attn.k_proj.weight",
"model.language_model.layers.30.self_attn.v_proj.weight",
"model.language_model.layers.30.mlp.gate_proj.weight",
"model.language_model.layers.30.mlp.up_proj.weight",
"model.language_model.layers.31.self_attn.q_proj.weight",
"model.language_model.layers.31.self_attn.k_proj.weight",
"model.language_model.layers.31.self_attn.v_proj.weight",
"model.language_model.layers.31.mlp.gate_proj.weight",
"model.language_model.layers.31.mlp.up_proj.weight",
"model.language_model.layers.32.self_attn.q_proj.weight",
"model.language_model.layers.32.self_attn.k_proj.weight",
"model.language_model.layers.32.self_attn.v_proj.weight",
"model.language_model.layers.32.mlp.gate_proj.weight",
"model.language_model.layers.32.mlp.up_proj.weight",
"model.language_model.layers.33.self_attn.q_proj.weight",
"model.language_model.layers.33.self_attn.k_proj.weight",
"model.language_model.layers.33.self_attn.v_proj.weight",
"model.language_model.layers.33.mlp.gate_proj.weight",
"model.language_model.layers.33.mlp.up_proj.weight",
"model.language_model.layers.34.self_attn.q_proj.weight",
"model.language_model.layers.34.self_attn.k_proj.weight",
"model.language_model.layers.34.self_attn.v_proj.weight",
"model.language_model.layers.34.mlp.gate_proj.weight",
"model.language_model.layers.34.mlp.up_proj.weight",
"model.language_model.layers.35.self_attn.q_proj.weight",
"model.language_model.layers.35.self_attn.k_proj.weight",
"model.language_model.layers.35.self_attn.v_proj.weight",
"model.language_model.layers.35.mlp.gate_proj.weight",
"model.language_model.layers.35.mlp.up_proj.weight"
]
},
"modules_quant_config": {},
"modules_to_not_convert": [
"patch_embed",
"patch_emb",
".vid_out",
"final_layer2",
".y_embedder",
".vid_in",
".norm_out",
".final_layer",
".txt_out",
"x_embedder",
"model.language_model.layers.*.self_attn.o_proj.weight",
".emb_in",
".proj_out",
".txt_in",
".emb_out",
".img_in",
".context_embedder",
"t_embedder2",
"wte",
".img_out",
"model.language_model.layers.*.mlp.down_proj.weight",
".condition_embedder",
"patch_embedding",
"lm_head.weight",
".time_embed",
"time_text_embed",
"t_embedder1",
"lm_head",
".t_embedder",
"multi_modal_projector",
".x_embedder"
],
"non_blocking": true,
"quant_conv": false,
"quant_embedding": false,
"quant_method": "sdnq",
"quantization_device": "cuda",
"quantized_matmul_dtype": null,
"return_device": "cuda",
"sdnq_version": "0.1.8",
"svd_rank": 32,
"svd_steps": 8,
"use_dynamic_quantization": true,
"use_grad_ckpt": true,
"use_quantized_matmul": false,
"use_quantized_matmul_conv": false,
"use_static_quantization": true,
"use_stochastic_rounding": false,
"use_svd": false,
"weights_dtype": "uint4"
},
"model_storage": {
"parameter_storage_bytes": 10597967456,
"parameter_storage_gib": 9.870126336812973
},
"quantization_peak_memory": {
"allocated_gib": 9.9013032913208,
"max_allocated_gib": 16.53543758392334,
"max_reserved_gib": 16.826171875,
"reserved_gib": 16.826171875
},
"quantization_seconds": 0.39238934498280287,
"repo_id": "WaveCut/HiDream-O1-Image-Dev-SDNQ-4bit-dynamic-uint4-th1e-2",
"sdnq_group_counts": {
"128": 4,
"269": 27,
"64": 265
},
"sdnq_layer_counts": {
"int5": 31,
"uint4": 265
},
"sdnq_layer_samples": [
{
"dtype": "int5",
"group_size": 128,
"name": "model.visual.blocks.0.attn.qkv",
"svd_rank": null
},
{
"dtype": "uint4",
"group_size": 64,
"name": "model.visual.blocks.0.attn.proj",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 128,
"name": "model.visual.blocks.0.mlp.linear_fc1",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 269,
"name": "model.visual.blocks.0.mlp.linear_fc2",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 128,
"name": "model.visual.blocks.1.attn.qkv",
"svd_rank": null
},
{
"dtype": "uint4",
"group_size": 64,
"name": "model.visual.blocks.1.attn.proj",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 128,
"name": "model.visual.blocks.1.mlp.linear_fc1",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 269,
"name": "model.visual.blocks.1.mlp.linear_fc2",
"svd_rank": null
},
{
"dtype": "uint4",
"group_size": 64,
"name": "model.visual.blocks.2.attn.qkv",
"svd_rank": null
},
{
"dtype": "uint4",
"group_size": 64,
"name": "model.visual.blocks.2.attn.proj",
"svd_rank": null
},
{
"dtype": "uint4",
"group_size": 64,
"name": "model.visual.blocks.2.mlp.linear_fc1",
"svd_rank": null
},
{
"dtype": "int5",
"group_size": 269,
"name": "model.visual.blocks.2.mlp.linear_fc2",
"svd_rank": null
}
],
"sdnq_param_counts": {
"int5": 151750656,
"uint4": 4949600256
},
"sdnq_storage_bytes": {
"int5": 97393536,
"uint4": 3093500160
},
"sdnq_storage_gib": {
"int5": 0.09070479869842529,
"uint4": 2.8810465335845947
},
"source_model": "HiDream-ai/HiDream-O1-Image-Dev",
"source_snapshot": "833d408a57a7c1e399757c7f2f174670726fd43c",
"variant": "pub_dynamic_uint4_th1e2_fixed"
}