Downloads · 30 days
29
41% of all-time downloads
huluhuluu/qwen3-4b-instruct-2507-executorch-qnn-htp
qwen3-4b-instruct-2507-executorch-qnn-htp is a machine learning model from huluhuluu. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for executorch.
This repository contains the ExecuTorch export artifacts used for the mobile benchmark runs on Qualcomm Snapdragon HTP devices. The model is Qwen3-4B-Instruct-2507.
Downloads · 30 days
29
41% of all-time downloads
All-time downloads
70
Public
Repo size
14.6 GB
Likes
0
Public
Click a slice to open those files.
.pth8 GB · 55%
From the Hugging Face model README
This repository contains the ExecuTorch export artifacts used for the mobile
benchmark runs on Qualcomm Snapdragon HTP devices. The model is
Qwen3-4B-Instruct-2507.
25609728363281281519361e-650000001282048temperature=0)sm8650/hybrid_llama_qnn.pte: ExecuTorch QNN/HTP artifact for Qualcomm
SM8650 (HTP v75-class devices).sm8750/qwen3_4b_hybrid_sm8750.pte: ExecuTorch QNN/HTP artifact for
Qualcomm SM8750 (HTP v79-class devices).sm8650/qwen3-4b-meta.pth: consolidated Meta-format checkpoint used as
the input to the SoC-specific export. It is not directly executable on a
phone.The PTE files are hardware-specific and should not be exchanged between SM8650 and SM8750 devices.
sm8650/qwen3-4b-meta.pth:
08eeaad348f70f17d2b6681b79f64f1c024dcc458e58b06c70722fc7dbde7dddsm8650/hybrid_llama_qnn.pte:
5d85935429607e636cc87077be7397d88837bc568e8adfaeb0d970b0f0e29a2dsm8750/qwen3_4b_hybrid_sm8750.pte:
78a2d703aebdf15568240d6e11ac8434d83936ee86554ddff4abf6a945e2a331