LLM EXPLORER 66,524 MODELS INDEXED

QwenPU by divagr1925

By divagr1925 · 769 downloads
HF Repo ↗ Copy URL

QwenPU is an open-source language model by divagr1925. Features: 4b LLM, VRAM: 26.6GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.36.

Base model:merge:qwen/qwen3.5-...   Base model:qwen/qwen3.5-4b   Compiled-weights   Conversational   Endpoints compatible   Gated-deltanet   Npu   Qwen3.5   Qwen3 5 text   Qwenputer   Region:us   Risc-v   Rv32im   Safetensors   Sharded   Tensorflow   Vector-unit
Model Card on HF 🤗: https://huggingface.co/divagr1925/QwenPU 

QwenPU Parameters and Internals

LLM NameQwenPU
Repository 🤗https://huggingface.co/divagr1925/QwenPU 
Base Model(s)  Qwen/Qwen3.5-4B   Qwen/Qwen3.5-4B
Model Size4b
Required VRAM26.6 GB
Updated2026-11-03
Maintainerdivagr1925
Model Typeqwen3_5_text
Model Files  6.6 GB: 1-of-11   2.2 GB: 2-of-11   2.2 GB: 3-of-11   2.2 GB: 4-of-11   2.2 GB: 5-of-11   2.2 GB: 6-of-11   2.2 GB: 7-of-11   2.2 GB: 8-of-11   2.2 GB: 9-of-11   2.2 GB: 10-of-11   0.2 GB: 11-of-11
Model ArchitectureQwen3_5ForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size642496
Errorsreplace

Best Alternatives to QwenPU

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.5 4B Qwenputer256K / 18.8 GB57297
Agents A1 4B Fable Preview256K / 8.4 GB1030
NeoHorse 1 4B256K / 8.4 GB1771179
Decider 4B256K / 8.4 GB2356517
...wen3.5 4B Qwenputer Filesystem256K / 22.4 GB68411
JevK5256K / 8.4 GB552513
Qwenputer Npu256K / 19.6 GB2643
Plumb 4B256K / 8.4 GB28256
JSBAI Coder 4B256K / 8.4 GB27951
Aztec Coder 4B256K / 8.4 GB11974
Note: green Score (e.g. "73.2") means that the model is better than divagr1925/QwenPU.