LLM EXPLORER 61,652 MODELS INDEXED

LongShuV3 Qwen3.5 122B A10B GameDev MLX 4bit by 5E1Labs

By 5E1Labs · 205 downloads
HF Repo ↗ Copy URL

LongShuV3 Qwen3.5 122B A10B GameDev MLX 4bit is an open-source language model by 5E1Labs. Features: 122b LLM, VRAM: 44.4GB, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.24.

  4-bit   4bit   Code-assistant   Conversational   En   Game-development   Gamedev   Gamedev-assistant   Instruct   Mlx   Mlx-lm   Moe   Multi-agent   Quantized   Qwen3 5 moe   Region:us   Roleplay   Safetensors   Sharded   Tensorflow   Unity   Unreal-engine   Zh
Model Card on HF 🤗: https://huggingface.co/5E1Labs/LongShuV3-Qwen3.5-122B-A10B-GameDev-MLX-4bit 

LongShuV3 Qwen3.5 122B A10B GameDev MLX 4bit Parameters and Internals

LLM NameLongShuV3 Qwen3.5 122B A10B GameDev MLX 4bit
Repository 🤗https://huggingface.co/5E1Labs/LongShuV3-Qwen3.5-122B-A10B-GameDev-MLX-4bit 
Base Model(s)  Qwen/Qwen3.5-122B-A10B-Instruct   Qwen/Qwen3.5-122B-A10B-Instruct
Model Size122b
Required VRAM44.4 GB
Updated2026-09-20
Maintainer5E1Labs
Model Typeqwen3_5_moe
Instruction-BasedYes
Model Files  5.2 GB: 1-of-9   5.3 GB: 2-of-9   5.4 GB: 3-of-9   5.1 GB: 4-of-9   5.3 GB: 5-of-9   5.2 GB: 6-of-9   5.3 GB: 7-of-9   5.2 GB: 8-of-9   2.4 GB: 9-of-9
Supported Languageszh en
Quantization Type4bit
Model ArchitectureQwen3_5MoeForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.5.0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>
Errorsreplace