LLM EXPLORER 61,755 MODELS INDEXED

Qwen3.8 35B A3B Distill OQ8 Fp16 Mtp by NovaeonStudio

By NovaeonStudio · 621 downloads
HF Repo ↗ Copy URL

Qwen3.8 35B A3B Distill OQ8 Fp16 Mtp is an open-source language model by NovaeonStudio. Features: 35b LLM, VRAM: 39.3GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.33.

  8-bit   Agentic   Apple-silicon Base model:empero-ai/qwen3.8-3... Base model:quantized:empero-ai...   Conversational   Distillation   En   Fp16   Function-calling   Image-text-to-text   Mlx   Moe   Novaeon   Omlx   Quantized   Qwen3 5 moe   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Tool-use
Model Card on HF 🤗: https://huggingface.co/NovaeonStudio/Qwen3.8-35B-A3B-Distill-oQ8-fp16-mtp 

Qwen3.8 35B A3B Distill OQ8 Fp16 Mtp Parameters and Internals

LLM NameQwen3.8 35B A3B Distill OQ8 Fp16 Mtp
Repository 🤗https://huggingface.co/NovaeonStudio/Qwen3.8-35B-A3B-Distill-oQ8-fp16-mtp 
Base Model(s)  empero-ai/Qwen3.8-35B-A3B-Distill   empero-ai/Qwen3.8-35B-A3B-Distill
Model Size35b
Required VRAM39.3 GB
Updated2026-09-20
MaintainerNovaeonStudio
Model Typeqwen3_5_moe
Model Files  5.3 GB: 1-of-8   5.1 GB: 2-of-8   5.1 GB: 3-of-8   5.1 GB: 4-of-8   5.1 GB: 5-of-8   5.1 GB: 6-of-8   5.1 GB: 7-of-8   3.4 GB: 8-of-8
Supported Languagesen
Quantization Typefp16
Model ArchitectureQwen3_5MoeForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.17.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Qwen3.8 35B A3B Distill OQ8 Fp16 Mtp

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.6 35B A3B MLX 4bit0K / 20.5 GB5131936
Qwen3.6 35B A3B MLX 6bit0K / 28.9 GB4785922
Qwen3.6 35B A3B MLX 8bit0K / 37.8 GB4735371
Qwen3.6 35B A3B UD MLX 4bit0K / 21.6 GB15720112
Qwen3.6 35B A3B OQ4e Fp16 Mtp0K / 22.4 GB1871
Qwen3.6 35B A3B MLX 8bit0K / 37.8 GB386241
Qwen3.6 35B A3B 2Bit GSQ0K / 13.5 GB21197
Qwen3.6 35B A3B 6bit0K / 28.9 GB27428
Qwen3.6 35B A3B UD MLX 3bit0K / 17.3 GB229213
Nail Qwen3.6 35B A3B MLX 5bit0K / 27.5 GB831
Note: green Score (e.g. "73.2") means that the model is better than NovaeonStudio/Qwen3.8-35B-A3B-Distill-oQ8-fp16-mtp.