Downloads · 30 days
2
13% of all-time downloads
qwen3.5-4b-lite-jax
gokashi/qwen3.5-4b-lite-jax
qwen3.5-4b-lite-jax is a machine learning model from gokashi. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This model is a LiteRT (TFLite) export of a Qwen 3.5 Hybrid model.
All-time downloads
15
Public
Repo size
17 B
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other1.5 KB · 81%
At a glance
- Access
- Public
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026
- SHA
- 2ba2d4e3
Model card
From the Hugging Face model README
gokashi/qwen3.5-4b-lite-jax
This model is a LiteRT (TFLite) export of a Qwen 3.5 Hybrid model.
Architecture
- Hybrid Layout: 3x Gated DeltaNet layers + 1x Gated Attention layer.
- Vocab Size: 248,320
- Target: Mobile/Edge Inference via LiteRT.
Usage
Run via TensorFlow Lite Interpreter with SELECT_TF_OPS enabled.
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026