LeanQuant commited on
Commit
cdd716c
·
verified ·
1 Parent(s): 611c7d3

Add files using upload-large-folder tool

Browse files
README.md ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - Wan-AI/Wan2.1-T2V-14B-Diffusers
4
+ base_model_relation: quantized
5
+ pipeline_tag: text-to-image
6
+ tags:
7
+ - dfloat11
8
+ - df11
9
+ - lossless compression
10
+ - 70% size, 100% accuracy
11
+ ---
12
+
13
+ # DFloat11 Compressed Model: `Wan-AI/Wan2.1-T2V-14B-Diffusers`
14
+
15
+ This model uses **DFloat11** lossless compression. It's 30% smaller than the original BFloat16 model, yet produces bit-identical outputs and runs efficiently on GPUs.
16
+
17
+ ### 📊 Performance Comparison
18
+
19
+ | Metric | Wan2.1-T2V-14B (BFloat16) | Wan2.1-T2V-14B (DFloat11) |
20
+ | ---------------------------------- | ------------------------- | ------------------------- |
21
+ | Model Size | 28.64 GB | 19.39 GB |
22
+ | Peak GPU Memory<br>(2s 480p Video) | 30.79 GB | 22.22 GB |
23
+ | Generation Time<br>(an A100 GPU) | 339 seconds | 348 seconds |
24
+
25
+ ### 🔍 How It Works
26
+
27
+ We apply Huffman coding to the exponent bits of BFloat16 model weights, which are highly compressible. We leverage hardware-aware algorithmic designs to enable highly efficient, on-the-fly weight decompression directly on the GPU. Find out more in our [research paper](https://arxiv.org/abs/2504.11651).
28
+
29
+ ### 🔧 How to Use
30
+
31
+ A complete usage guide is available in our GitHub repository: [https://github.com/LeanModels/DFloat11/tree/master/examples/wan2.1](https://github.com/LeanModels/DFloat11/tree/master/examples/wan2.1).
32
+
33
+ ### 📄 Learn More
34
+
35
+ * **Paper**: [70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float](https://arxiv.org/abs/2504.11651)
36
+ * **GitHub**: [https://github.com/LeanModels/DFloat11](https://github.com/LeanModels/DFloat11)
37
+ * **HuggingFace**: [https://huggingface.co/DFloat11](https://huggingface.co/DFloat11)
blocks_0.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd8bfcfc5d60bb0d1d28b7ed01f8900c426b8e8b7a2a09d6b52724b0cb629de2
3
+ size 469822461
blocks_1.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0ad5a049edb0987bdc7bddad07970ac4ead7fe34e9a51832f5ca5f0fcd31c87
3
+ size 473112003
blocks_10.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:068472fc20c4682bc291b2c3aa3a612140ea51c26a806dc05bee2ae53ac364af
3
+ size 474336115
blocks_11.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdd677d4afea88bf4681c58b2c6072aa6922245849cf78f5b9c044731da45a74
3
+ size 474120463
blocks_12.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3d1c74188cda421419fb187ad8a25b79770bd9c4d50235202c17d0644c084f0
3
+ size 473831508
blocks_13.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f12b244606ef5f7b0bde47b3e3a96a2e10693b70a9682dd8e0b22eec8387533d
3
+ size 473374022
blocks_14.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42b429b7acd2d2b119d81463a06519b419af642a999314d275455ac1ef4f2a46
3
+ size 473103178
blocks_15.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8aa54771d6dadf9aab31cb1b6912856a786c97d1642f599957fc7cd5900143a5
3
+ size 472704290
blocks_16.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98718faba517a25df92da135e00ba882be7245d4156ed7097284bddaeb638a1f
3
+ size 472829300
blocks_17.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:985220b7be09bf72cc52765a845d29e06e6970c3a5019d1ba77683cbaa5a6f2d
3
+ size 472559745
blocks_18.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:226eb04924cc67da299e8609d2e77171c3d83bbac3889dc06e6d29882796a729
3
+ size 472655400
blocks_19.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f7a86bcc9eb7dd20b3c033541d9de14a8870e75fed0162c54f09734c774360c
3
+ size 472160049
blocks_2.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3974d8643be38e3cfca88e9cd4c7516d929ca3014c1980909d1852381bd8a351
3
+ size 474056136
blocks_20.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa4f5890bf0d62efecc0f90d0757817315066c0249b2c79570af1fd4856c8214
3
+ size 471888174
blocks_21.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:552b6903820d4c05a0a1e18c951e70df425f40cbd352b2a230eb3b459911449e
3
+ size 471952382
blocks_22.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5087b44aa92c7533fc65f194568d47891c8ce27e54a61a73b4e777a7a2e591
3
+ size 471547890
blocks_23.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5313a6b33fe531d3acaedff7fbb8496b8c2959db08b1bb9f4d4aeec98698270a
3
+ size 471574444
blocks_24.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcfb54e90efbeffd163cdbb66fb1f338f2a8a59f7c26f9166aed497864d12c6b
3
+ size 471244483
blocks_25.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a06217caa5c0cda5a1ce23a6abc7b4df6c7fc2b6163249eb96e5c6a4ee38e2d
3
+ size 471059022
blocks_26.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a20e9947523259e8981fcfbba4a0b62660d29fb73d796d2dd2b5f8ffcee58b9
3
+ size 470851176
blocks_27.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c5f0d57ea63a62638fbec542e958fbc05c49b58a4eff649c99b37fdeb6a86bb
3
+ size 470701666
blocks_28.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5816b4d7478a13bc0f6bdb6fa468a2cb6c743e9e10130fb2ddd53ec344d75bc1
3
+ size 470648690
blocks_29.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8376a75503ef50412b88eb1a4ddb04155e182c8f5c50f1e1112ee620b7d6a7c4
3
+ size 470299716
blocks_3.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e30f23b73be0484efdbd164ae33da18c3d491ba0e2e58f33e7b206be36e2bcb
3
+ size 474198955
blocks_30.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f815dd0530ac2209e1518881335e265bd56664c0f54cbcf64b5997ae7258f96
3
+ size 470266634
blocks_31.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08e7ee4a1823549fbdfe7d071e1fb9135db22d94b0410c55f74d1ddd9dc59922
3
+ size 470037585
blocks_32.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:863b6264f585b7e6b47f5d9decf506211f8666cd86b12b18f21f703b968927c1
3
+ size 469643405
blocks_33.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2568a4de725d66d6b74d8f9d5d817a21e1e14d643ab9bf7de7b424408bd6362
3
+ size 469162662
blocks_34.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de978cb26fd4c374e21efe920be8ab2ae3281618150a968f7ef04a09c77bce1b
3
+ size 469284418
blocks_35.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5aacd685673224543da4f7dea1777b5d63712e598297113d4f9635ceee191786
3
+ size 469715909
blocks_36.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f16985e60efa0a471ca88382482e1788ef9394e4bd6193be8079eb24d1771d5c
3
+ size 469094034
blocks_37.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcb6e50b4a7df9d0a6e61acbaa7328dac72c51b4fccc9c02abf9281016801e31
3
+ size 468906382
blocks_38.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ef2537591f12ed2dd5b74400fa3763c062fe69d1c07617a9d1904fcbb922a9c
3
+ size 468289750
blocks_39.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:625e5841f2843e6a3f534b5e72979f6412da37e8f0555d815201854a2c11548a
3
+ size 468007542
blocks_4.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3698d798735dec37ce6581dd3dce54061df2c2540d1e84b25cca06263d790b5
3
+ size 474332041
blocks_5.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e98c92ceb29b204153d366cdd79a6b7a72929aa69286b1b9b0c791f5dd21e8cc
3
+ size 474445010
blocks_6.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d16154b435ca229e6db638a8adebda2fdc9ca4d857549bc14e81f1c0817e0e8
3
+ size 474545789
blocks_7.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:392c0ff1aa92a046b783e78f936d557b8e07783fa6be2db93cf4048217e8b527
3
+ size 474624395
blocks_8.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9193508c1834cf5d6187f024abd670eafaf3ab807ba41f8cea000f075eb4728b
3
+ size 474667732
blocks_9.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35238929d7fbe55af4b3be8903b2a8b46c4dd484ed9c2b9636c13e1757c1592b
3
+ size 474616125
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dfloat11_config": {
3
+ "bytes_per_thread": 8,
4
+ "pattern_dict": {
5
+ "blocks\\.\\d+": [
6
+ "attn1.to_q",
7
+ "attn1.to_k",
8
+ "attn1.to_v",
9
+ "attn1.to_out.0",
10
+ "attn2.to_q",
11
+ "attn2.to_k",
12
+ "attn2.to_v",
13
+ "attn2.to_out.0",
14
+ "ffn.net.0.proj",
15
+ "ffn.net.2"
16
+ ]
17
+ },
18
+ "threads_per_block": [
19
+ 512
20
+ ],
21
+ "version": "0.2.0"
22
+ },
23
+ "model_type": "llama"
24
+ }
diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95d6132378b2ec2434463e628fef7741c76b5a811807e937bd5f757b69eebfb8
3
+ size 520531712