LLM EXPLORER 58,608 MODELS INDEXED

Qwen3.8 27B GPTQ W4A16 by pearsonkyle

By pearsonkyle · 126 downloads
HF Repo ↗ Copy URL

Qwen3.8 27B GPTQ W4A16 is an open-source language model by pearsonkyle. Features: 27b LLM, VRAM: 0.8GB, License: apache-2.0, Quantized.

  27b   32k-calibration   4-bit   Agentic Base model:quantized:qwen/qwen...   Base model:qwen/qwen3.8-27b   Compressed-tensors   Conversational   En   Endpoints compatible   Function-calling   Gptq   Image-text-to-text   Int4   Llm-compressor   Long-context   Mtp   Multimodal   Quantization   Quantized   Qwen   Qwen3   Qwen3 5   Region:us   Safetensors   Speculative-decoding   Tool-use   Vision   Vllm   W4a16
Model Card on HF 🤗: https://huggingface.co/pearsonkyle/Qwen3.8-27B-GPTQ-W4A16 

Qwen3.8 27B GPTQ W4A16 Parameters and Internals

LLM NameQwen3.8 27B GPTQ W4A16
Repository 🤗https://huggingface.co/pearsonkyle/Qwen3.8-27B-GPTQ-W4A16 
Base Model(s)  Qwen/Qwen3.8-27B   Qwen/Qwen3.8-27B
Model Size27b
Required VRAM0.8 GB
Updated2026-08-17
Maintainerpearsonkyle
Model Typeqwen3_5
Model Files  0.8 GB   0.9 GB   17.6 GB
Supported Languagesen
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.8.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Qwen3.8 27B GPTQ W4A16

Best Alternatives
Context / RAM
Downloads
Likes
...en3.8 27B W4A16 AutoRound GPTQ0K / 19.2 GB37485
Qwen3.8 27B GPTQ 4bit0K / 21 GB24335
Qwen3.6 27B GPTQ Pro 4bit0K / 18.8 GB41540739
Qwen3.5 27B GPTQ Int40K / 30.1 GB10538157
... PristinelyUncensored GPTQ MTP0K / 19.6 GB121
...s3.5 27B V3 PolarQuant V7 GPTQ0K / 19.2 GB3023
Qwopus3.5 27B V3 HLWQ V7 GPTQ0K / 19.2 GB336
Qwen3.5 27B GPTQ Int40K / 28.5 GB241
...7B GPTQ Int4 Sym G128 MTP BF160K / 19.6 GB5363
Qwen3.8 27B GPTQ Int40K / 19.6 GB8811
Note: green Score (e.g. "73.2") means that the model is better than pearsonkyle/Qwen3.8-27B-GPTQ-W4A16.