LLM EXPLORER 66,845 MODELS INDEXED

Qwopus3.6 27B Coder NVFP4 by maci0

By maci0 · 505871 downloads
HF Repo ↗ Copy URL

Qwopus3.6 27B Coder NVFP4 is an open-source language model by maci0. Features: 27b LLM, VRAM: 1.8GB, License: apache-2.0, Code Generating, LLM Explorer Score: 0.35.

  8-bit Base model:jackrong/qwopus3.6-... Base model:quantized:jackrong/...   Code   Codegen   Coder   Compressed-tensors   Conversational   En   Endpoints compatible   Fp4   Gptq   Image-text-to-text   Llm-compressor   Nvfp4   Quantized   Qwen3 5   Region:us   Safetensors   Thinking   Vision   Vision-language   Vllm   W4a4   Zh
Model Card on HF 🤗: https://huggingface.co/maci0/Qwopus3.6-27B-Coder-NVFP4 

Qwopus3.6 27B Coder NVFP4 Parameters and Internals

LLM NameQwopus3.6 27B Coder NVFP4
Repository 🤗https://huggingface.co/maci0/Qwopus3.6-27B-Coder-NVFP4 
Model NameUnsloth/qwen3.6-27B
Base Model(s)  Jackrong/Qwopus3.6-27B-Coder   Jackrong/Qwopus3.6-27B-Coder
Model Size27b
Required VRAM1.8 GB
Updated2026-08-09
Maintainermaci0
Model Typeqwen3_5
Model Files  1.8 GB   18.8 GB
Supported Languagesen zh
Generates CodeYes
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Tokenizer ClassTokenizersBackend
Padding Token<|vision_pad|>
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwopus3.6 27B Coder NVFP4

Best Alternatives
Context / RAM
Downloads
Likes
Qwopus3.8 27B Coder Flash0K / 55.6 GB65228
Industry Coder 27B V2.00K / 55.6 GB1122
... Coder Flash EXL3 3.50bpw 24GB0K / 15 GB871
....8 27B Coder Flash Exl3 4.0bpw0K / 16.4 GB552
Qwen3.8 27B USS Prism0K / 55.5 GB680
Qwen3.8 27B CodeForge 3xMerge0K / 55.5 GB490
Veriloop Coder E10K / 55.6 GB80725
Qwimi 3.6 27B Coder MTP BF160K / 55.4 GB964
Damork Coder 27B Multimodal V20K / 54.7 GB880
Damork Coder 27B Multimodal V20K / 54.7 GB90
Note: green Score (e.g. "73.2") means that the model is better than maci0/Qwopus3.6-27B-Coder-NVFP4.