LLM EXPLORER 58,873 MODELS INDEXED

Qwen3.6 27B Bnb 4bit by lokeshe09

By lokeshe09 · 1194 downloads
HF Repo ↗ Copy URL

Qwen3.6 27B Bnb 4bit is an open-source language model by lokeshe09. Features: 27b LLM, VRAM: 17.7GB, Context: 256K, Quantized, LLM Explorer Score: 0.26.

  Arxiv:1910.09700   4-bit   4bit   Bitsandbytes   Conversational   Endpoints compatible   Quantized   Qwen3 5 text   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/lokeshe09/Qwen3.6-27B-bnb-4bit 

Qwen3.6 27B Bnb 4bit Parameters and Internals

LLM NameQwen3.6 27B Bnb 4bit
Repository 🤗https://huggingface.co/lokeshe09/Qwen3.6-27B-bnb-4bit 
Model Size27b
Required VRAM17.7 GB
Updated2026-08-17
Maintainerlokeshe09
Model Typeqwen3_5_text
Model Files  2.5 GB: 1-of-9   2.5 GB: 2-of-9   2.0 GB: 3-of-9   2.0 GB: 4-of-9   2.0 GB: 5-of-9   2.0 GB: 6-of-9   2.0 GB: 7-of-9   2.0 GB: 8-of-9   0.7 GB: 9-of-9
Quantization Type4bit
Model ArchitectureQwen3_5ForCausalLM
Context Length262144
Model Max Length262144
Transformers Version5.6.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size248320
Errorsreplace

Best Alternatives to Qwen3.6 27B Bnb 4bit

Best Alternatives
Context / RAM
Downloads
Likes
...en3.6 27B OBLITERATED MLX 4bit256K / 15.1 GB6463
Qwen3.8 27B Abliteratex256K / 53.9 GB3216
Qwen3.6 27B OBLITERATED256K / 53.6 GB15137184
Qwen3.8 Whittle 16B256K / 2.4 GB17191
...wen3.8 P44w75 16.8B Unrepaired256K / 2.4 GB9291
...maAQUA Gridbook 13GB 5080 Vllm256K / 13 GB2872
Focus Red256K / 53.8 GB4161
...2 Abliterated OptiQ 3.7bpw Mlx256K / 13.4 GB672714
...2 Abliterated OptiQ 3.7bpw Mlx256K / 13.4 GB591914
Qwen3.6 27B Esper4256K / 55.6 GB29219
Note: green Score (e.g. "73.2") means that the model is better than lokeshe09/Qwen3.6-27B-bnb-4bit.