Hugging Face's logo

EdgeFirst
/
yolov8-det

Object Detection
LiteRT
ONNX
edgefirst
edge-ai
npu
int8
yolo
nxp
hailo
jetson
embedded
Eval Results (legacy)
Model card Files
xet
 Community
1
sebastientaylor commited on
Commit
98c1d16
Β·
verified Β·
1 Parent(s): 6cbd706

Update model card for yolov8-det

Browse files
Files changed (1) hide show
  1. README.md +34 -7
README.md CHANGED
@@ -34,7 +34,7 @@ model-index:
34
  value: 59.4
35
  - name: "mAP@0.5-0.95 (Small ONNX FP32)"
36
  type: map
37
- value: 43.22
38
  - name: "mAP@0.5 (Medium ONNX FP32)"
39
  type: map_50
40
  value: 64.68
@@ -64,9 +64,9 @@ Accuracy ceiling for each size, measured against COCO `val2017` (5,000 images) w
64
 
65
  | Size | Params | GFLOPs | mAP@0.5 | mAP@0.5-0.95 | mAP@0.75 | Source |
66
  |------|--------|--------|---------|--------------|----------|--------|
67
- | Nano | 3.2M | 8.9 | 50.49% | 35.83% | 38.84% | [v-a88](https://edgefirst.studio/public/validation/v-a88/details?mode=charts) |
68
- | Small | 11.2M | 28.8 | 59.40% | 43.22% | 46.68% | [v-a8f](https://edgefirst.studio/public/validation/v-a8f/details?mode=charts) |
69
- | Medium | 25.9M | 79.3 | 64.68% | 48.45% | 52.62% | [v-a96](https://edgefirst.studio/public/validation/v-a96/details?mode=charts) |
70
  | Large | 43.7M | 165.7 | β€” | β€” | β€” | β€” |
71
  | XLarge | 68.2M | 258.5 | β€” | β€” | β€” | β€” |
72
 
@@ -105,10 +105,19 @@ Row conventions in the table below:
105
 
106
  | Size | Platform | mAP@0.5 | Ξ” vs FP32 (pp) | mAP@0.5-0.95 | Inference (ms) | End-to-end (ms) | Realized FPS | Core-throughput ceiling (FPS) | Source |
107
  |------|----------|---------|----------------|--------------|----------------|-----------------|--------------|-------------------------------|--------|
 
 
 
108
  | Nano | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 50.49% | +0.00 | 35.83% | 38.01 | 49.96 | 71.6 | ~72 | [v-a46](https://edgefirst.studio/public/validation/v-a46/details?mode=charts) |
109
- | Nano | ONNX FP32 (CUDA) | 50.49% | ref | 35.83% | 8.04 | 13.22 | 350.5 | ~350 | [v-a88](https://edgefirst.studio/public/validation/v-a88/details?mode=charts) |
 
 
 
110
  | Nano | ONNX FP16 (CUDA) | 50.49% | +0.00 | 35.79% | 6.23 | 11.70 | 457.3 | ~457 | [v-a9d](https://edgefirst.studio/public/validation/v-a9d/details?mode=charts) |
111
  | Nano | ONNX FP32 | 50.49% | +0.00 | 35.82% | 37.75 | 55.49 | 96.0 | ~27 | [v-cd2](https://edgefirst.studio/public/validation/v-cd2/details?mode=charts) |
 
 
 
112
  | Nano | Apple M2 Max β€” CoreML Neural Engine (FP16) | 50.12% | -0.37 | 35.52% | 1.80 | 5.26 | 791.0 | ~744 | [v-9cd](https://edgefirst.studio/public/validation/v-9cd/details?mode=charts) |
113
  | Nano | Apple M2 Max β€” CoreML Metal GPU (FP16) | 50.13% | -0.36 | 35.52% | 6.00 | 9.73 | 445.0 | ~445 | [v-9ce](https://edgefirst.studio/public/validation/v-9ce/details?mode=charts) |
114
  | Nano | Apple M2 Max β€” CoreML CPU (FP16) | 50.14% | -0.35 | 35.50% | 15.93 | 19.96 | 117.5 | ~117 | [v-9cf](https://edgefirst.studio/public/validation/v-9cf/details?mode=charts) |
@@ -131,10 +140,19 @@ Row conventions in the table below:
131
  | Nano | NXP Ara240 (FRDM) β€” throughput | 46.63% | -3.86 | 32.13% | 6.87 | 49.38 | 210.8 | ~298 | [v-a13](https://edgefirst.studio/public/validation/v-a13/details?mode=charts) |
132
  | Nano | Raspberry Pi 5 + Hailo-8L NPU | 49.80% | -0.69 | 35.22% | 13.20 | 24.86 | 64.9 | ~65 | [v-8df](https://edgefirst.studio/public/validation/v-8df/details?mode=charts) |
133
  | Nano | NVIDIA Jetson Orin Nano (TensorRT FP16) | 50.50% | +0.01 | 35.79% | 11.93 | 26.90 | 259.7 | ~259 | [v-91a](https://edgefirst.studio/public/validation/v-91a/details?mode=charts) |
 
 
 
134
  | Small | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 59.40% | +0.00 | 43.22% | 94.66 | 106.27 | 29.8 | ~30 | [v-a4d](https://edgefirst.studio/public/validation/v-a4d/details?mode=charts) |
135
- | Small | ONNX FP32 (CUDA) | 59.40% | ref | 43.22% | 15.48 | 20.84 | 204.5 | ~204 | [v-a8f](https://edgefirst.studio/public/validation/v-a8f/details?mode=charts) |
 
 
 
136
  | Small | ONNX FP16 (CUDA) | 59.37% | -0.03 | 43.19% | 10.64 | 16.31 | 299.3 | ~299 | [v-aa4](https://edgefirst.studio/public/validation/v-aa4/details?mode=charts) |
137
  | Small | ONNX FP32 | 59.37% | -0.03 | 43.18% | 73.55 | 83.57 | 54.0 | ~14 | [v-cda](https://edgefirst.studio/public/validation/v-cda/details?mode=charts) |
 
 
 
138
  | Small | Apple M2 Max β€” CoreML Neural Engine (FP16) | 58.80% | -0.60 | 42.78% | 5.01 | 8.46 | 342.6 | ~344 | [v-9ed](https://edgefirst.studio/public/validation/v-9ed/details?mode=charts) |
139
  | Small | Apple M2 Max β€” CoreML Metal GPU (FP16) | 58.82% | -0.58 | 42.73% | 15.93 | 19.52 | 178.9 | ~179 | [v-9ee](https://edgefirst.studio/public/validation/v-9ee/details?mode=charts) |
140
  | Small | Apple M2 Max β€” CoreML CPU (FP16) | 58.83% | -0.57 | 42.70% | 32.00 | 36.09 | 60.1 | ~60 | [v-9ef](https://edgefirst.studio/public/validation/v-9ef/details?mode=charts) |
@@ -158,10 +176,19 @@ Row conventions in the table below:
158
  | Small | NXP Ara240 (FRDM) β€” throughput | 56.70% | -2.70 | 39.81% | 10.27 | 24.17 | 142.4 | ~142 | [v-a21](https://edgefirst.studio/public/validation/v-a21/details?mode=charts) |
159
  | Small | Raspberry Pi 5 + Hailo-8L NPU | 58.52% | -0.88 | 42.34% | 30.36 | 45.07 | 28.8 | ~29 | [v-8ec](https://edgefirst.studio/public/validation/v-8ec/details?mode=charts) |
160
  | Small | NVIDIA Jetson Orin Nano (TensorRT FP16) | 59.38% | -0.02 | 43.15% | 24.62 | 37.12 | 157.5 | ~157 | [v-922](https://edgefirst.studio/public/validation/v-922/details?mode=charts) |
 
 
 
161
  | Medium | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 64.68% | +0.00 | 48.44% | 243.22 | 255.15 | 11.8 | ~12 | [v-a54](https://edgefirst.studio/public/validation/v-a54/details?mode=charts) |
162
- | Medium | ONNX FP32 (CUDA) | 64.68% | ref | 48.45% | 43.49 | 49.46 | 78.5 | ~78 | [v-a96](https://edgefirst.studio/public/validation/v-a96/details?mode=charts) |
 
 
 
163
  | Medium | ONNX FP16 (CUDA) | 64.68% | +0.00 | 48.40% | 23.13 | 29.44 | 147.7 | ~148 | [v-aab](https://edgefirst.studio/public/validation/v-aab/details?mode=charts) |
164
  | Medium | ONNX FP32 | 64.70% | +0.02 | 48.46% | 258.20 | 265.50 | 3.9 | ~4 | [v-cdb](https://edgefirst.studio/public/validation/v-cdb/details?mode=charts) |
 
 
 
165
  | Medium | Apple M2 Max β€” CoreML Neural Engine (FP16) | 63.82% | -0.86 | 47.68% | 13.41 | 17.27 | 140.9 | ~142 | [v-75e](https://edgefirst.studio/public/validation/v-75e/details?mode=charts) |
166
  | Medium | Apple M2 Max β€” CoreML Metal GPU (FP16) | 63.82% | -0.86 | 47.70% | 37.93 | 42.15 | 77.0 | ~77 | [v-768](https://edgefirst.studio/public/validation/v-768/details?mode=charts) |
167
  | Medium | Apple M2 Max β€” CoreML CPU (FP16) | 63.87% | -0.81 | 47.68% | 61.61 | 65.71 | 31.8 | ~32 | [v-9dd](https://edgefirst.studio/public/validation/v-9dd/details?mode=charts) |
 
34
  value: 59.4
35
  - name: "mAP@0.5-0.95 (Small ONNX FP32)"
36
  type: map
37
+ value: 43.21
38
  - name: "mAP@0.5 (Medium ONNX FP32)"
39
  type: map_50
40
  value: 64.68
 
64
 
65
  | Size | Params | GFLOPs | mAP@0.5 | mAP@0.5-0.95 | mAP@0.75 | Source |
66
  |------|--------|--------|---------|--------------|----------|--------|
67
+ | Nano | 3.2M | 8.9 | 50.49% | 35.83% | 38.83% | [v-e89](https://edgefirst.studio/public/validation/v-e89/details?mode=charts) |
68
+ | Small | 11.2M | 28.8 | 59.40% | 43.21% | 46.68% | [v-e8d](https://edgefirst.studio/public/validation/v-e8d/details?mode=charts) |
69
+ | Medium | 25.9M | 79.3 | 64.68% | 48.45% | 52.62% | [v-e8e](https://edgefirst.studio/public/validation/v-e8e/details?mode=charts) |
70
  | Large | 43.7M | 165.7 | β€” | β€” | β€” | β€” |
71
  | XLarge | 68.2M | 258.5 | β€” | β€” | β€” | β€” |
72
 
 
105
 
106
  | Size | Platform | mAP@0.5 | Ξ” vs FP32 (pp) | mAP@0.5-0.95 | Inference (ms) | End-to-end (ms) | Realized FPS | Core-throughput ceiling (FPS) | Source |
107
  |------|----------|---------|----------------|--------------|----------------|-----------------|--------------|-------------------------------|--------|
108
+ | Nano | ONNX FP32 (AWS Graviton Β· 4-core) | 50.54% | +0.05 | 35.85% | 273.95 | 283.13 | 14.4 | ~4 | [v-e57](https://edgefirst.studio/public/validation/v-e57/details?mode=charts) |
109
+ | Nano | ONNX FP32 (AWS Graviton4 Β· 8-core) | 50.54% | +0.05 | 35.85% | 232.86 | 239.57 | 33.9 | ~4 | [v-e6d](https://edgefirst.studio/public/validation/v-e6d/details?mode=charts) |
110
+ | Nano | ONNX FP32 (AWS Graviton4 Β· 48-core) | 50.54% | +0.05 | 35.86% | 79.20 | 91.32 | 157.6 | ~13 | [v-e69](https://edgefirst.studio/public/validation/v-e69/details?mode=charts) |
111
  | Nano | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 50.49% | +0.00 | 35.83% | 38.01 | 49.96 | 71.6 | ~72 | [v-a46](https://edgefirst.studio/public/validation/v-a46/details?mode=charts) |
112
+ | Nano | ONNX FP32 (Intel Xeon Platinum 8488C Β· 24-core) | 50.49% | +0.00 | 35.83% | 74.88 | 96.12 | 148.5 | ~13 | [v-e84](https://edgefirst.studio/public/validation/v-e84/details?mode=charts) |
113
+ | Nano | ONNX FP32 (Intel Xeon Platinum 8488C Β· 4-core) | 50.49% | +0.00 | 35.82% | 136.60 | 147.24 | 55.3 | ~7 | [v-e78](https://edgefirst.studio/public/validation/v-e78/details?mode=charts) |
114
+ | Nano | ONNX FP32 (CUDA) | 50.49% | ref | 35.83% | 7.32 | 16.39 | 319.8 | ~320 | [v-e89](https://edgefirst.studio/public/validation/v-e89/details?mode=charts) |
115
+ | Nano | ONNX FP32 (CUDA) | 50.49% | +0.00 | 35.83% | 8.04 | 13.22 | 350.5 | ~350 | [v-a88](https://edgefirst.studio/public/validation/v-a88/details?mode=charts) |
116
  | Nano | ONNX FP16 (CUDA) | 50.49% | +0.00 | 35.79% | 6.23 | 11.70 | 457.3 | ~457 | [v-a9d](https://edgefirst.studio/public/validation/v-a9d/details?mode=charts) |
117
  | Nano | ONNX FP32 | 50.49% | +0.00 | 35.82% | 37.75 | 55.49 | 96.0 | ~27 | [v-cd2](https://edgefirst.studio/public/validation/v-cd2/details?mode=charts) |
118
+ | Nano | ONNX FP32 | 50.54% | +0.05 | 35.85% | 29.85 | 40.21 | 130.3 | ~34 | [v-cd0](https://edgefirst.studio/public/validation/v-cd0/details?mode=charts) |
119
+ | Nano | ONNX FP32 | 50.54% | +0.05 | 35.85% | 39.30 | 46.42 | 25.2 | ~26 | [v-ccc](https://edgefirst.studio/public/validation/v-ccc/details?mode=charts) |
120
+ | Nano | ONNX FP32 | 50.49% | +0.00 | 35.82% | 39.49 | 48.59 | 23.1 | ~26 | [v-ccb](https://edgefirst.studio/public/validation/v-ccb/details?mode=charts) |
121
  | Nano | Apple M2 Max β€” CoreML Neural Engine (FP16) | 50.12% | -0.37 | 35.52% | 1.80 | 5.26 | 791.0 | ~744 | [v-9cd](https://edgefirst.studio/public/validation/v-9cd/details?mode=charts) |
122
  | Nano | Apple M2 Max β€” CoreML Metal GPU (FP16) | 50.13% | -0.36 | 35.52% | 6.00 | 9.73 | 445.0 | ~445 | [v-9ce](https://edgefirst.studio/public/validation/v-9ce/details?mode=charts) |
123
  | Nano | Apple M2 Max β€” CoreML CPU (FP16) | 50.14% | -0.35 | 35.50% | 15.93 | 19.96 | 117.5 | ~117 | [v-9cf](https://edgefirst.studio/public/validation/v-9cf/details?mode=charts) |
 
140
  | Nano | NXP Ara240 (FRDM) β€” throughput | 46.63% | -3.86 | 32.13% | 6.87 | 49.38 | 210.8 | ~298 | [v-a13](https://edgefirst.studio/public/validation/v-a13/details?mode=charts) |
141
  | Nano | Raspberry Pi 5 + Hailo-8L NPU | 49.80% | -0.69 | 35.22% | 13.20 | 24.86 | 64.9 | ~65 | [v-8df](https://edgefirst.studio/public/validation/v-8df/details?mode=charts) |
142
  | Nano | NVIDIA Jetson Orin Nano (TensorRT FP16) | 50.50% | +0.01 | 35.79% | 11.93 | 26.90 | 259.7 | ~259 | [v-91a](https://edgefirst.studio/public/validation/v-91a/details?mode=charts) |
143
+ | Small | ONNX FP32 (AWS Graviton Β· 4-core) | 59.37% | -0.03 | 43.18% | 800.24 | 809.83 | 5.0 | ~1 | [v-e5a](https://edgefirst.studio/public/validation/v-e5a/details?mode=charts) |
144
+ | Small | ONNX FP32 (AWS Graviton4 Β· 8-core) | 59.37% | -0.03 | 43.18% | 657.27 | 663.61 | 12.1 | ~2 | [v-e75](https://edgefirst.studio/public/validation/v-e75/details?mode=charts) |
145
+ | Small | ONNX FP32 (AWS Graviton4 Β· 48-core) | 59.37% | -0.03 | 43.18% | 216.01 | 227.07 | 59.0 | ~5 | [v-e70](https://edgefirst.studio/public/validation/v-e70/details?mode=charts) |
146
  | Small | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 59.40% | +0.00 | 43.22% | 94.66 | 106.27 | 29.8 | ~30 | [v-a4d](https://edgefirst.studio/public/validation/v-a4d/details?mode=charts) |
147
+ | Small | ONNX FP32 (Intel Xeon Platinum 8488C Β· 24-core) | 59.40% | +0.00 | 43.21% | 180.67 | 202.24 | 68.3 | ~6 | [v-e6e](https://edgefirst.studio/public/validation/v-e6e/details?mode=charts) |
148
+ | Small | ONNX FP32 (Intel Xeon Platinum 8488C Β· 4-core) | 59.40% | +0.00 | 43.22% | 414.03 | 424.50 | 19.1 | ~2 | [v-e77](https://edgefirst.studio/public/validation/v-e77/details?mode=charts) |
149
+ | Small | ONNX FP32 (CUDA) | 59.40% | ref | 43.21% | 11.25 | 20.20 | 300.1 | ~300 | [v-e8d](https://edgefirst.studio/public/validation/v-e8d/details?mode=charts) |
150
+ | Small | ONNX FP32 (CUDA) | 59.40% | +0.00 | 43.22% | 15.48 | 20.84 | 204.5 | ~204 | [v-a8f](https://edgefirst.studio/public/validation/v-a8f/details?mode=charts) |
151
  | Small | ONNX FP16 (CUDA) | 59.37% | -0.03 | 43.19% | 10.64 | 16.31 | 299.3 | ~299 | [v-aa4](https://edgefirst.studio/public/validation/v-aa4/details?mode=charts) |
152
  | Small | ONNX FP32 | 59.37% | -0.03 | 43.18% | 73.55 | 83.57 | 54.0 | ~14 | [v-cda](https://edgefirst.studio/public/validation/v-cda/details?mode=charts) |
153
+ | Small | ONNX FP32 | 59.37% | -0.03 | 43.18% | 103.65 | 110.89 | 9.6 | ~10 | [v-cd6](https://edgefirst.studio/public/validation/v-cd6/details?mode=charts) |
154
+ | Small | ONNX FP32 | 59.40% | +0.00 | 43.21% | 65.52 | 73.86 | 14.7 | ~15 | [v-cd5](https://edgefirst.studio/public/validation/v-cd5/details?mode=charts) |
155
+ | Small | ONNX FP32 | 59.40% | +0.00 | 43.21% | 70.28 | 89.27 | 54.1 | ~14 | [v-cd3](https://edgefirst.studio/public/validation/v-cd3/details?mode=charts) |
156
  | Small | Apple M2 Max β€” CoreML Neural Engine (FP16) | 58.80% | -0.60 | 42.78% | 5.01 | 8.46 | 342.6 | ~344 | [v-9ed](https://edgefirst.studio/public/validation/v-9ed/details?mode=charts) |
157
  | Small | Apple M2 Max β€” CoreML Metal GPU (FP16) | 58.82% | -0.58 | 42.73% | 15.93 | 19.52 | 178.9 | ~179 | [v-9ee](https://edgefirst.studio/public/validation/v-9ee/details?mode=charts) |
158
  | Small | Apple M2 Max β€” CoreML CPU (FP16) | 58.83% | -0.57 | 42.70% | 32.00 | 36.09 | 60.1 | ~60 | [v-9ef](https://edgefirst.studio/public/validation/v-9ef/details?mode=charts) |
 
176
  | Small | NXP Ara240 (FRDM) β€” throughput | 56.70% | -2.70 | 39.81% | 10.27 | 24.17 | 142.4 | ~142 | [v-a21](https://edgefirst.studio/public/validation/v-a21/details?mode=charts) |
177
  | Small | Raspberry Pi 5 + Hailo-8L NPU | 58.52% | -0.88 | 42.34% | 30.36 | 45.07 | 28.8 | ~29 | [v-8ec](https://edgefirst.studio/public/validation/v-8ec/details?mode=charts) |
178
  | Small | NVIDIA Jetson Orin Nano (TensorRT FP16) | 59.38% | -0.02 | 43.15% | 24.62 | 37.12 | 157.5 | ~157 | [v-922](https://edgefirst.studio/public/validation/v-922/details?mode=charts) |
179
+ | Medium | ONNX FP32 (AWS Graviton Β· 4-core) | 64.70% | +0.02 | 48.46% | 2671.61 | 2687.79 | 1.5 | ~0 | [v-e5c](https://edgefirst.studio/public/validation/v-e5c/details?mode=charts) |
180
+ | Medium | ONNX FP32 (AWS Graviton4 Β· 48-core) | 64.70% | +0.02 | 48.46% | 552.57 | 563.48 | 23.1 | ~2 | [v-e85](https://edgefirst.studio/public/validation/v-e85/details?mode=charts) |
181
+ | Medium | ONNX FP32 (AWS Graviton4 Β· 8-core) | 64.70% | +0.02 | 48.46% | 1706.42 | 1712.42 | 4.7 | ~1 | [v-e6c](https://edgefirst.studio/public/validation/v-e6c/details?mode=charts) |
182
  | Medium | ONNX FP32 (Intel Core i9-13900F Β· 32-core) | 64.68% | +0.00 | 48.44% | 243.22 | 255.15 | 11.8 | ~12 | [v-a54](https://edgefirst.studio/public/validation/v-a54/details?mode=charts) |
183
+ | Medium | ONNX FP32 (Intel Xeon Platinum 8488C Β· 24-core) | 64.68% | +0.00 | 48.44% | 438.65 | 459.61 | 28.3 | ~2 | [v-e68](https://edgefirst.studio/public/validation/v-e68/details?mode=charts) |
184
+ | Medium | ONNX FP32 (Intel Xeon Platinum 8488C Β· 4-core) | 64.68% | +0.00 | 48.44% | 962.24 | 971.48 | 8.3 | ~1 | [v-e7b](https://edgefirst.studio/public/validation/v-e7b/details?mode=charts) |
185
+ | Medium | ONNX FP32 (CUDA) | 64.68% | ref | 48.45% | 23.84 | 31.81 | 157.3 | ~157 | [v-e8e](https://edgefirst.studio/public/validation/v-e8e/details?mode=charts) |
186
+ | Medium | ONNX FP32 (CUDA) | 64.68% | +0.00 | 48.45% | 43.49 | 49.46 | 78.5 | ~78 | [v-a96](https://edgefirst.studio/public/validation/v-a96/details?mode=charts) |
187
  | Medium | ONNX FP16 (CUDA) | 64.68% | +0.00 | 48.40% | 23.13 | 29.44 | 147.7 | ~148 | [v-aab](https://edgefirst.studio/public/validation/v-aab/details?mode=charts) |
188
  | Medium | ONNX FP32 | 64.70% | +0.02 | 48.46% | 258.20 | 265.50 | 3.9 | ~4 | [v-cdb](https://edgefirst.studio/public/validation/v-cdb/details?mode=charts) |
189
+ | Medium | ONNX FP32 | 64.68% | +0.00 | 48.44% | 159.36 | 178.18 | 24.4 | ~6 | [v-cd9](https://edgefirst.studio/public/validation/v-cd9/details?mode=charts) |
190
+ | Medium | ONNX FP32 | 64.70% | +0.02 | 48.46% | 178.38 | 188.38 | 22.4 | ~6 | [v-cd7](https://edgefirst.studio/public/validation/v-cd7/details?mode=charts) |
191
+ | Medium | ONNX FP32 | 64.68% | +0.00 | 48.44% | 150.86 | 158.90 | 6.5 | ~7 | [v-cd4](https://edgefirst.studio/public/validation/v-cd4/details?mode=charts) |
192
  | Medium | Apple M2 Max β€” CoreML Neural Engine (FP16) | 63.82% | -0.86 | 47.68% | 13.41 | 17.27 | 140.9 | ~142 | [v-75e](https://edgefirst.studio/public/validation/v-75e/details?mode=charts) |
193
  | Medium | Apple M2 Max β€” CoreML Metal GPU (FP16) | 63.82% | -0.86 | 47.70% | 37.93 | 42.15 | 77.0 | ~77 | [v-768](https://edgefirst.studio/public/validation/v-768/details?mode=charts) |
194
  | Medium | Apple M2 Max β€” CoreML CPU (FP16) | 63.87% | -0.81 | 47.68% | 61.61 | 65.71 | 31.8 | ~32 | [v-9dd](https://edgefirst.studio/public/validation/v-9dd/details?mode=charts) |