LLM EXPLORER 66,845 MODELS INDEXED

Qwen3.5 2B by Qwen

By Qwen · 2633311 downloads
HF Repo ↗ Copy URL

Qwen3.5 2B is an open-source language model by Qwen. Features: 2b LLM, VRAM: 4.5GB, License: apache-2.0, LLM Explorer Score: 0.39.

Base model:finetune:qwen/qwen3... Base model:qwen/qwen3.5-2b-bas...   Conversational   Deploy:azure   Deploy:sagemaker   Endpoints compatible   Eval-results   Image-text-to-text   Qwen3 5   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3.5-2B 

Qwen3.5 2B Parameters and Internals

LLM NameQwen3.5 2B
Repository 🤗https://huggingface.co/Qwen/Qwen3.5-2B 
Base Model(s)  Qwen/Qwen3.5-2B-Base   Qwen/Qwen3.5-2B-Base
Model Size2b
Required VRAM4.5 GB
Updated2026-08-08
MaintainerQwen
Model Typeqwen3_5
Model Files  4.5 GB: 1-of-1
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version4.57.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Quantized Models of the Qwen3.5 2B

Model
Likes
Downloads
VRAM
Ternary Bonsai 2 2B Mlx 2bit1132551 GB
...B Uncensored Zero Refusal SOMA225591 GB
Qwen3.5 2B Hermes Meta Ads12822 GB
Qwen3.5 2B Heretic016051 GB
Synapse Alpha 1.0 4bit1431 GB
Qwen3.5 2B MLX 4bit2123101 GB
Mavery Geselle 2B MLX 4bit2771 GB
Qwen3.5 2B OptiQ 4bit1212960 GB
Spartan 1.2 2B MLX 8bit0271 GB

Best Alternatives to Qwen3.5 2B

Best Alternatives
Context / RAM
Downloads
Likes
MiniCPM OCR Jev0K / 4.4 GB177080
Index Translate 2B0K / 4.5 GB208965
RynnBrain1.1 2B0K / 4.4 GB26857912
Qwen3.5 2B Base0K / 4.5 GB80959182
WeMM Embedding 2B0K / 5.4 GB543368
Qwen3.8 2B Distill Heretic V20K / 4.4 GB8362
GLiDE 2B0K / 3.8 GB1094
...e Qwen3.5 2B Decision MLX Bf160K / 3.8 GB28012
Qwen3.5 2B0K / 4.5 GB10348718
Wang Qwen3.5 2B DocTags ONNX0K /  GB541
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3.5-2B.