Downloads · 30 days
67
25% of all-time downloads
Tensor-0.1-35B-A3B
Akushon/Tensor-0.1-35B-A3B
Tensor-0.1-35B-A3B is a text generation model from Akushon. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as other.
A Self-postrained NVFP4-quantized 35B MoE model derived from Qwen3.5-35B-A3B.
All-time downloads
266
Public
Parameters
17.9B
38.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors21 GB · 100%
Parameter types
How the weights are stored.
U816.8B · 94%
Try a prompt
Base models
Model card
From the Hugging Face model README
Tensor-0.1-35B-A3B
A Self-postrained NVFP4-quantized 35B MoE model derived from Qwen3.5-35B-A3B.
Overview
| Property | Value |
|---|---|
| Base model | Qwen3.5-35B-A3B |
| Architecture | Qwen3_5Moe (Mixture-of-Experts) |
| Quantization | NVFP4 (4-bit floating point) |
| Context length | 262,144 tokens |
| Model size | ~17 GB |
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("Akushon/Tensor-0.1-35B-A3B")
tokenizer = AutoTokenizer.from_pretrained("Akushon/Tensor-0.1-35B-A3B")
Note: This model requires NVFP4 quantization support. Use a compatible NVFP4-aware inference backend.
License
This model is released under the same license terms as the base model.
- Task
- Text Generation
- Library
- transformers
- Type
- qwen3_5_moe_text
- License
- other
- Languages
- en, zh
- Created
- Aug 30, 2026
- Updated
- Sep 2, 2026