LLM EXPLORER 58,608 MODELS INDEXED

Qwen3.8 27B Heretic Ara OQ4e Mtp by underlotus

By underlotus · 915 downloads
HF Repo ↗ Copy URL

Qwen3.8 27B Heretic Ara OQ4e Mtp is an open-source language model by underlotus. Features: 27b LLM, VRAM: 16.9GB, License: apache-2.0, LLM Explorer Score: 0.35.

  4-bit   4bit   Abliterated   Apple-silicon Base model:quantized:trohrbaug... Base model:trohrbaugh/qwen3.8-...   Conversational   En   Image-text-to-text   Mixed-precision   Mlx   Mtp   Oq   Qwen   Qwen3 5   Region:us   Safetensors   Sharded   Tensorflow   Uncensored   Zh
Model Card on HF 🤗: https://huggingface.co/underlotus/Qwen3.8-27B-heretic-ara-oQ4e-mtp 

Qwen3.8 27B Heretic Ara OQ4e Mtp Parameters and Internals

LLM NameQwen3.8 27B Heretic Ara OQ4e Mtp
Repository 🤗https://huggingface.co/underlotus/Qwen3.8-27B-heretic-ara-oQ4e-mtp 
Base Model(s)  trohrbaugh/Qwen3.8-27B-heretic-ara   trohrbaugh/Qwen3.8-27B-heretic-ara
Model Size27b
Required VRAM16.9 GB
Updated2026-08-16
Maintainerunderlotus
Model Typeqwen3_5
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   1.9 GB: 4-of-4
Supported Languagesen zh
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.14.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Qwen3.8 27B Heretic Ara OQ4e Mtp

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.8 27B FP80K / 0.4 GB123157416
Qwen3.8 27B0K / 55.6 GB28753
Qwen3.6 27B0K / 55.4 GB67281382197
Qwen3.6 27B NVFP40K / 22 GB1887525419
...n3.8 27B MTPLX Optimized Speed0K / 19.6 GB698950
ThinkingCap Qwen3.6 27B0K / 54.7 GB35947614
Qwen3.6 27B FP80K / 0.4 GB7740472342
Qwen3.8 27B NVFP40K / 22 GB3012214
Qwen3.5 27B0K / 55.2 GB25362941028
Qwen3.8 27B INT8 W8A16 MTP0K / 30.7 GB472326
Note: green Score (e.g. "73.2") means that the model is better than underlotus/Qwen3.8-27B-heretic-ara-oQ4e-mtp.