LLM EXPLORER 59,694 MODELS INDEXED

Dirk Qwen3.8 27B MLX OQ6e by peculiar-ragdoll

By peculiar-ragdoll · 121 downloads
HF Repo ↗ Copy URL

Dirk Qwen3.8 27B MLX OQ6e is an open-source language model by peculiar-ragdoll. Features: 27b LLM, VRAM: 23.6GB, License: apache-2.0, LLM Explorer Score: 0.29.

  6-bit   Agentic-coding   Apple-silicon Base model:quantized:qwen/qwen...   Base model:qwen/qwen3.8-27b   Conversational   En   Image-text-to-text   Mlx   Mtp   Omlx   Oq   Quantized   Qwen3.8   Qwen3 5   Region:us   Safetensors   Sharded   Speculative-decoding   Tensorflow   Token-efficient   Vision   Zh
Model Card on HF 🤗: https://huggingface.co/peculiar-ragdoll/Dirk-Qwen3.8-27B-MLX-oQ6e 

Dirk Qwen3.8 27B MLX OQ6e Parameters and Internals

LLM NameDirk Qwen3.8 27B MLX OQ6e
Repository 🤗https://huggingface.co/peculiar-ragdoll/Dirk-Qwen3.8-27B-MLX-oQ6e 
Base Model(s)  Qwen/Qwen3.8-27B   Qwen/Qwen3.8-27B
Model Size27b
Required VRAM23.6 GB
Updated2026-08-27
Maintainerpeculiar-ragdoll
Model Typeqwen3_5
Model Files  5.1 GB: 1-of-5   5.0 GB: 2-of-5   5.0 GB: 3-of-5   5.0 GB: 4-of-5   3.5 GB: 5-of-5
Supported Languagesen zh
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.8.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Dirk Qwen3.8 27B MLX OQ6e

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.8 27B MLX0K / 16 GB141118
Qwen3.8 27B0K / 55.6 GB272721
Qwen3.8 27B FP80K / 0.4 GB123157416
Qwen3.6 27B0K / 55.4 GB67281382197
Qwen3.8 27B0K / 55.6 GB28753
Qwen3.6 27B NVFP40K / 22 GB1887525419
Qwen3.6 27B FP80K / 0.4 GB7740472342
ThinkingCap Qwen3.6 27B0K / 54.7 GB35947614
Qwen3.5 27B0K / 55.2 GB25362941028
Qwen3.5 27B0K / 55.2 GB811516
Note: green Score (e.g. "73.2") means that the model is better than peculiar-ragdoll/Dirk-Qwen3.8-27B-MLX-oQ6e.