qaihm-bot commited on
Commit
e7831ac
·
verified ·
1 Parent(s): 3c09c11

See https://github.com/quic/ai-hub-models/releases/v0.34.0 for changelog.

README.md CHANGED
@@ -23,6 +23,7 @@ More details on model performance across various devices, can be found
23
  [here](https://aihub.qualcomm.com/models/fastsam_s).
24
 
25
 
 
26
  ### Model Details
27
 
28
  - **Model Type:** Model_use_case.semantic_segmentation
@@ -35,31 +36,31 @@ More details on model performance across various devices, can be found
35
 
36
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | FastSam-S | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 38.516 ms | 4 - 54 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
39
  | FastSam-S | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 38.22 ms | 2 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
40
- | FastSam-S | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 13.272 ms | 4 - 63 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
41
  | FastSam-S | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 15.382 ms | 5 - 47 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
42
- | FastSam-S | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.251 ms | 4 - 30 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
43
  | FastSam-S | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 7.148 ms | 5 - 21 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
44
- | FastSam-S | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.146 ms | 4 - 55 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
45
  | FastSam-S | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 11.012 ms | 2 - 73 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
46
- | FastSam-S | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 38.516 ms | 4 - 54 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
47
  | FastSam-S | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 38.22 ms | 2 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
48
- | FastSam-S | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 7.315 ms | 4 - 31 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
49
  | FastSam-S | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 7.187 ms | 0 - 26 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
50
- | FastSam-S | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.782 ms | 4 - 45 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
51
  | FastSam-S | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 13.562 ms | 5 - 41 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
52
- | FastSam-S | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 7.092 ms | 4 - 30 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
53
  | FastSam-S | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 7.158 ms | 0 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
54
- | FastSam-S | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.146 ms | 4 - 55 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
55
  | FastSam-S | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 11.012 ms | 2 - 73 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
56
- | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.052 ms | 4 - 31 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
57
  | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 7.133 ms | 0 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
58
  | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 8.43 ms | 0 - 70 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
59
- | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.433 ms | 3 - 67 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
60
  | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 5.34 ms | 5 - 104 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
61
  | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.415 ms | 15 - 115 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
62
- | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 4.836 ms | 0 - 55 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
63
  | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 3.997 ms | 5 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
64
  | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.863 ms | 15 - 99 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
65
  | FastSam-S | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 7.883 ms | 24 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
@@ -121,17 +122,7 @@ device. This script does the following:
121
  ```bash
122
  python -m qai_hub_models.models.fastsam_s.export
123
  ```
124
- ```
125
- Profiling Results
126
- ------------------------------------------------------------
127
- FastSam-S
128
- Device : cs_8275 (ANDROID 14)
129
- Runtime : TFLITE
130
- Estimated inference time (ms) : 38.5
131
- Estimated peak memory usage (MB): [4, 54]
132
- Total # Ops : 287
133
- Compute Unit(s) : npu (287 ops) gpu (0 ops) cpu (0 ops)
134
- ```
135
 
136
 
137
  ## How does this work?
 
23
  [here](https://aihub.qualcomm.com/models/fastsam_s).
24
 
25
 
26
+
27
  ### Model Details
28
 
29
  - **Model Type:** Model_use_case.semantic_segmentation
 
36
 
37
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
+ | FastSam-S | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 38.563 ms | 4 - 54 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
40
  | FastSam-S | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 38.22 ms | 2 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
41
+ | FastSam-S | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 13.254 ms | 4 - 59 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
42
  | FastSam-S | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 15.382 ms | 5 - 47 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
43
+ | FastSam-S | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 7.306 ms | 4 - 31 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
44
  | FastSam-S | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 7.148 ms | 5 - 21 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
45
+ | FastSam-S | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.155 ms | 5 - 55 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
46
  | FastSam-S | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 11.012 ms | 2 - 73 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
47
+ | FastSam-S | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 38.563 ms | 4 - 54 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
48
  | FastSam-S | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 38.22 ms | 2 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
49
+ | FastSam-S | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 7.28 ms | 4 - 30 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
50
  | FastSam-S | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 7.187 ms | 0 - 26 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
51
+ | FastSam-S | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.778 ms | 4 - 45 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
52
  | FastSam-S | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 13.562 ms | 5 - 41 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
53
+ | FastSam-S | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 7.307 ms | 4 - 30 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
54
  | FastSam-S | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 7.158 ms | 0 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
55
+ | FastSam-S | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.155 ms | 5 - 55 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
56
  | FastSam-S | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 11.012 ms | 2 - 73 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
57
+ | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 7.335 ms | 4 - 30 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
58
  | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 7.133 ms | 0 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
59
  | FastSam-S | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 8.43 ms | 0 - 70 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
60
+ | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 5.454 ms | 3 - 73 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
61
  | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 5.34 ms | 5 - 104 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
62
  | FastSam-S | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.415 ms | 15 - 115 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
63
+ | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 4.832 ms | 3 - 58 MB | NPU | [FastSam-S.tflite](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.tflite) |
64
  | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 3.997 ms | 5 - 83 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
65
  | FastSam-S | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.863 ms | 15 - 99 MB | NPU | [FastSam-S.onnx](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.onnx) |
66
  | FastSam-S | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 7.883 ms | 24 - 24 MB | NPU | [FastSam-S.dlc](https://huggingface.co/qualcomm/FastSam-S/blob/main/FastSam-S.dlc) |
 
122
  ```bash
123
  python -m qai_hub_models.models.fastsam_s.export
124
  ```
125
+
 
 
 
 
 
 
 
 
 
 
126
 
127
 
128
  ## How does this work?
precompiled/qualcomm-snapdragon-x-elite/FastSam-S.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:eb85a1219d7a760b9c1ae7d1f32c9d84d4530adc7fc063bc72844255073312c2
3
- size 22348142
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15a3cb086960ca9df7cf67bbab52dad37751671579d8cc617a085d15a02884e4
3
+ size 22348113
precompiled/qualcomm-snapdragon-x-elite/sdk_versions.yml ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ sdk_versions:
2
+ qnn_context_binary:
3
+ qairt: 2.34.2.250528164111_119506
4
+ precompiled_qnn_onnx:
5
+ qairt: 2.33.2.250410134701_117956