LLM EXPLORER 62,323 MODELS INDEXED

Qwen3 4B Diffusion Base by fwerkor

By fwerkor · 222 downloads

Qwen3 4B Diffusion Base is an open-source language model by fwerkor. Features: 4b LLM, VRAM: 8.9GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.31.

  Arxiv:2608.10438 Base model:finetune:qwen/qwen3...   Base model:qwen/qwen3-4b-base   Bidirectional   Cid Continuous-interaction-diffusi...   Conversational   Custom code   Dataset:openbmb/ultra-fineweb   Dataset:openbmb/ultradata-code   Dataset:openbmb/ultradata-math   Dataset:openbmb/ultrax-preview   Diffusion-language-model   Endpoints compatible   Masked-diffusion   Qwen3   Region:us   Research   Safetensors   Sharded   Tensorflow

Qwen3 4B Diffusion Base Parameters and Internals

LLM NameQwen3 4B Diffusion Base
Repository πŸ€—https://huggingface.co/fwerkor/Qwen3-4B-Diffusion-Base 
Base Model(s)  Qwen3 4B Base   Qwen/Qwen3-4B-Base
Model Size4b
Required VRAM8.9 GB
Updated2026-09-27
Maintainerfwerkor
Model Typeqwen3
Model Files  4.0 GB: 1-of-3   4.0 GB: 2-of-3   0.9 GB: 3-of-3
Model ArchitectureCIDDiffusionForMaskedLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.57.6
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Errorsreplace