AXERA-TECH / EdgeTAM

huggingface.co
Total runs: 68
24-hour runs: 4
7-day runs: 8
30-day runs: 63
Model's Last Updated: January 30 2026
image-segmentation

Introduction of EdgeTAM

Model Details of EdgeTAM

EdgeTAM

基于EdgeSAM的图像分割Pipeline,支持多种输入提示(框、点、掩码),支持650N系列平台的模型推理。

支持芯片:

  • AX650N

支持硬件

原始模型请参考

性能对比
  • 输入图片大小 512x512
Models Latency (ms) CMM Usage (MB)
edgetam_image_encoder 22.348 29.124
edgetam_prompt_encoder 0.055 0.023
edgetam_prompt_mask_encoder 0.457 0.037
edgetam_mask_decoder 4.729 16.730
模型转换
  • 模型转换工具链 Pulsar2
  • 转换文档[TODO]
环境准备

安装需要的python库 pip install -r requirements.txt

运行
(myenv) root@ax650:~/EdgeTAM# python3 image_prediction_ax.py --input_box 75,275,1725,850
[INFO] Available providers:  ['AxEngineExecutionProvider']
Loading EdgeTAM Onnx models...
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Chip type: ChipType.MC50
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Engine version: 2.12.0s
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
Get prompts:
  input_box: [  75  275 1725  850]
  input_point_coords: None
  input_point_labels: None
Only box input provided
Get dense_embeddings_no_mask
[0.9777304]
✅ Saved: ./results/mask_1.png

保存结果在 ./results 目录下: image

(myenv) root@ax650:~/EdgeTAM# python3 image_prediction_ax.py --image_path ./examples/images/truck.jpg --input_box 425,600,700,875 --input_point_coords 575,750 --input_point_labels 0
[INFO] Available providers:  ['AxEngineExecutionProvider']
Loading EdgeTAM Onnx models...
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Chip type: ChipType.MC50
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Engine version: 2.12.0s
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 5.0-patch1-dirty a512c95e-dirty
['575,750']
575,750
Get prompts:
  input_box: [425 600 700 875]
  input_point_coords: [[575 750]]
  input_point_labels: [0]
Get dense_embeddings_no_mask
[0.90291053]
✅ Saved: ./results/mask_1.png

image

Runs of AXERA-TECH EdgeTAM on huggingface.co

68
Total runs
4
24-hour runs
4
3-day runs
8
7-day runs
63
30-day runs

More Information About EdgeTAM huggingface.co Model

More EdgeTAM license Visit here:

https://choosealicense.com/licenses/afl-3.0

EdgeTAM huggingface.co

EdgeTAM huggingface.co is an AI model on huggingface.co that provides EdgeTAM's model effect (), which can be used instantly with this AXERA-TECH EdgeTAM model. huggingface.co supports a free trial of the EdgeTAM model, and also provides paid use of the EdgeTAM. Support call EdgeTAM model through api, including Node.js, Python, http.

AXERA-TECH EdgeTAM online free

EdgeTAM huggingface.co is an online trial and call api platform, which integrates EdgeTAM's modeling effects, including api services, and provides a free online trial of EdgeTAM, you can try EdgeTAM online for free by clicking the link below.

AXERA-TECH EdgeTAM online free url in huggingface.co:

https://huggingface.co/AXERA-TECH/EdgeTAM

EdgeTAM install

EdgeTAM is an open source model from GitHub that offers a free installation service, and any user can find EdgeTAM on GitHub to install. At the same time, huggingface.co provides the effect of EdgeTAM install, users can directly use EdgeTAM installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

EdgeTAM install url in huggingface.co:

https://huggingface.co/AXERA-TECH/EdgeTAM

Url of EdgeTAM

Provider of EdgeTAM huggingface.co

AXERA-TECH
ORGANIZATIONS

Other API from AXERA-TECH

huggingface.co

Total runs: 100
Run Growth: 45
Growth Rate: 45.00%
Updated:March 09 2026
huggingface.co

Total runs: 94
Run Growth: 66
Growth Rate: 70.21%
Updated:October 24 2025
huggingface.co

Total runs: 83
Run Growth: 52
Growth Rate: 62.65%
Updated:January 15 2026
huggingface.co

Total runs: 79
Run Growth: 51
Growth Rate: 64.56%
Updated:November 25 2025
huggingface.co

Total runs: 78
Run Growth: 46
Growth Rate: 58.97%
Updated:January 21 2026
huggingface.co

Total runs: 75
Run Growth: 25
Growth Rate: 33.33%
Updated:November 02 2025
huggingface.co

Total runs: 71
Run Growth: -101
Growth Rate: -142.25%
Updated:February 13 2026
huggingface.co

Total runs: 70
Run Growth: 59
Growth Rate: 84.29%
Updated:October 08 2025
huggingface.co

Total runs: 70
Run Growth: 41
Growth Rate: 62.12%
Updated:January 23 2026
huggingface.co

Total runs: 69
Run Growth: 61
Growth Rate: 93.85%
Updated:January 30 2026
huggingface.co

Total runs: 67
Run Growth: 59
Growth Rate: 88.06%
Updated:November 25 2025
huggingface.co

Total runs: 65
Run Growth: 59
Growth Rate: 90.77%
Updated:September 24 2025
huggingface.co

Total runs: 63
Run Growth: 59
Growth Rate: 100.00%
Updated:December 30 2025
huggingface.co

Total runs: 63
Run Growth: 63
Growth Rate: 100.00%
Updated:October 20 2025
huggingface.co

Total runs: 34
Run Growth: -5
Growth Rate: -14.71%
Updated:August 04 2025
huggingface.co

Total runs: 26
Run Growth: -11
Growth Rate: -42.31%
Updated:August 04 2025
huggingface.co

Total runs: 21
Run Growth: 21
Growth Rate: 100.00%
Updated:October 25 2025
huggingface.co

Total runs: 12
Run Growth: -9
Growth Rate: -75.00%
Updated:March 23 2026
huggingface.co

Total runs: 11
Run Growth: 7
Growth Rate: 63.64%
Updated:December 10 2025
huggingface.co

Total runs: 9
Run Growth: 4
Growth Rate: 44.44%
Updated:January 23 2026