Skip to content
ModelListedExternal
Browse files

YOLOv8 Pose

This version of YOLOv8-POSE has been converted to run on the Axera NPU using w8a16 quantization.

Sourcehuggingface.co/AXERA-TECH/YOLOv8-Pose/tree/480ece640bc1b8340397c744c8187cb779f5be1f
KindOtherBase modelUltralytics/YOLOv8VersionRevision 480ece6Bgrade

Model source

Kind
Other
File size
12.6–265 MB
Base model
Ultralytics/YOLOv8
Version
vRevision 480ece6
Updated
Oct 21, 2026
Downloads
218
Source
Hugging Face

Source description

YOLOv8-POSE

This version of YOLOv8-POSE has been converted to run on the Axera NPU using w8a16 quantization.

This model has been optimized with the following LoRA:

Compatible with Pulsar2 version: 5.0

Convert tools links:

For those who are interested in model conversion, you can try to export axmodel through

Support Platform

Performance Statistics

AX650N

ModelLatency(ms) npu1Latency(ms) npu3
yolov8n-pose4.1351.453
yolov8s-pose10.8423.764
yolov8m-pose27.5159.577
yolov8l-pose52.56018.221
yolov8x-pose90.25729.313

AX630C

ModelLatency(ms) npu1Latency(ms) npu2
yolov8n-pose14.7388.213
yolov8s-pose33.15724.109
yolov8m-pose82.26757.532
yolov8l-pose145.686101.811
yolov8x-pose248.331160.857

AX615

ModelLatency(ms) npu1Latency(ms) npu2
yolov8n-pose17.40210.083
yolov8s-pose50.24028.091
yolov8m-pose136.38670.737

AX637

ModelLatency(ms) npu1
yolov8n-pose4.838
yolov8s-pose12.250
yolov8m-pose28.938
yolov8l-pose55.146
yolov8x-pose86.095

How to use

Download all files from this repository to the device

Inference

Input image:

Inference with AX650 Host, such as M4N-Dock(爱芯派Pro)

(base) root@ax650:~/ax650pose# python3 ax_infer.py --model-path yolov8s-pose_640x640_npu3.axmodel --test-img bus.jpg
[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: 6.0 a498e20d
[YOLOv8-Pose] [15:35:43.230] [DEBUG] Load model time = 576.32 ms
[YOLOv8-Pose] [15:35:43.301] [DEBUG] Pre-process time = 21.58 ms
[YOLOv8-Pose] [15:35:43.328] [DEBUG] Forward time = 26.25 ms
[YOLOv8-Pose] [15:35:43.335] [DEBUG] Post-process time = 6.46 ms
[YOLOv8-Pose] [15:35:43.341] [INFO] Draw Results (4 persons):
[YOLOv8-Pose] [15:35:43.342] [INFO] (222, 404, 344, 859) -> person: 0.89
[YOLOv8-Pose] [15:35:43.347] [INFO] (49, 397, 244, 902) -> person: 0.88
[YOLOv8-Pose] [15:35:43.348] [INFO] (669, 392, 808, 875) -> person: 0.87
[YOLOv8-Pose] [15:35:43.349] [INFO] (0, 413, 78, 935) -> person: 0.57
[YOLOv8-Pose] [15:35:43.377] [INFO] Saved to result_yolov8_pose.jpg

Output image:

Sources

1 source
Hugging FacePrimaryVersion 480ece640bc1b8340397c744c8187cb779f5be1f
huggingface.co/AXERA-TECH/YOLOv8-Pose/tree/480ece640bc1b8340397c744c8187cb779f5be1f

Verified Oct 21

Download files

Model artifacts

5 artifacts

yolov8l-pose_640x640.onnx

onnx · 170 MB · SHA-256 47070ae5e072…cc01 · Hugging Face

Download file

yolov8m-pose_640x640.onnx

onnx · 101 MB · SHA-256 9153dd5d5a9c…acde · Hugging Face

Download file

yolov8n-pose_640x640.onnx

onnx · 12.6 MB · SHA-256 c28854a5e413…da23 · Hugging Face

Download file

yolov8s-pose_640x640.onnx

onnx · 44.4 MB · SHA-256 50582ba675ec…2834 · Hugging Face

Download file

yolov8x-pose_640x640.onnx

onnx · 265 MB · SHA-256 25c64b8f7d38…ef70 · Hugging Face

Download file

Source excerpts

2 excerpts
ReadmeAug 5huggingface.co/AXERA-TECH/YOLOv8-Pose

--- license: mit language: - en base_model: - Ultralytics/YOLOv8 pipeline_tag: object-detection tags: - Ultralytics - YOLOv8 - YOLOv8-POSE --- # YOLOv8-POSE This version of YOLOv8-POSE has been converted to run on the Axera NPU using **w8a16** quantization. This model has been optimized with the following LoRA: Compatible with Pulsar2 version: 5.0 ## Convert tools links: For those who are interested in model conversion, you can try to export axmodel through - [The repo of ax-samples](https://github.com/AXERA-TECH/ax-samples), which you can get the how to build the `ax_yolov8_pose` - [The repo of axcl-samples](https://github.com/AXERA-TECH/axcl-samples), which you can get the how to build the `axcl_yolov8_pose` - [Pulsar2 Link, How to Convert ONNX to axmodel](https://pulsar2-docs.readthedocs.io/en/latest/pulsar2/introduction.html) ## Support Platform - **AX650N/AX8850** - [M4N-Dock(爱芯派Pro)](https://wiki.sipeed.com/hardware/zh/maixIV/m4ndock/m4ndock.html) - [M.2 Accelerator card](https://docs.m5stack.com/en/ai_hardware/LLM-8850_Card) - **AX630C** - [爱芯派2](https://axera-pi-2-docs-cn.readthedocs.io/zh-cn/latest/index.html) - [Module-LLM](https://docs.m5stack.com/zh_CN/module/Module-LLM) - [LLM630 Compute Kit](https://docs.m5stack.com/zh_CN/core/LLM630%20Compute%20Kit) - **AX615** - **AX637** ### Performance Statistics #### AX650N | Model | Latency(ms) npu1 | Latency(ms) npu3 | | :--- | :---: | :---: | | **yolov8n-pose** | 4.135 | 1.453 | | **yolov8s-pose** | 10.842 | 3.764 | | **yolov8m-pose** | 27.515 | 9.577 | | **yolov8l-pose** | 52.560 | 18.221 | | **yolov8x-pose** | 90.257 | 29.313 | #### AX630C | Model | Latency(ms) npu1 | Latency(ms) npu2 | | :--- | :---: | :---: | | **yolov8n-pose** | 14.738 | 8.213 | | **yolov8s-pose** | 33.157 | 24.109 | | **yolov8m-pose** | 82.267 | 57.532 | | **yolov8l-pose** | 145.686 | 101.811 | | **yolov8x-pose** | 248.331 | 160.857 | #### AX615 | Model | Latency(ms) npu1 | Latency(ms) npu2 | | :--- | :---: | :---: | | **yolov8n-pose** | 17.402 | 10.083 | | **yolov8s-pose** | 50.240 | 28.091 | | **yolov8m-pose** | 136.386 | 70.737 | #### AX637 | Model | Latency(ms) npu1 | | :--- | :---: | | **yolov8n-pose** | 4.838 | | **yolov8s-pose** | 12.250 | | **yolov8m-pose** | 28.938 | | **yolov8l-pose** | 55.146 | | **yolov8x-pose** | 86.095 | ## How to use Download all files from this repository to the device...

NoteAug 5huggingface.co/AXERA-TECH/YOLOv8-Pose

Source context: 218 downloads · 1 likes · Pipeline object-detection · Repo AXERA-TECH/YOLOv8-Pose