LLM EXPLORER 55,709 MODELS INDEXED

Gemma 4 12B Coder Fable5 Composer2.5 V1 OptiQ 4bit by mlx-community

By mlx-community · 2920 downloads
HF Repo ↗ Copy URL

Gemma 4 12B Coder Fable5 Composer2.5 V1 OptiQ 4bit is an open-source language model by mlx-community. Features: 12b LLM, VRAM: 8.8GB, License: apache-2.0, Quantized, Code Generating.

  4-bit   4bit   8bit   Apple-silicon Base model:quantized:yuxinlu1/... Base model:yuxinlu1/gemma-4-12...   Code   Codegen   Coding   Conversational   Gemma4   Gemma4 unified   Image-text-to-text   Mixed-precision   Mlx   Optiq   Quantized   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Thinking   Vision   Vision-language
Model Card on HF 🤗: https://huggingface.co/mlx-community/gemma-4-12B-coder-fable5-composer2.5-v1-OptiQ-4bit 

Gemma 4 12B Coder Fable5 Composer2.5 V1 OptiQ 4bit Parameters and Internals

LLM NameGemma 4 12B Coder Fable5 Composer2.5 V1 OptiQ 4bit
Repository 🤗https://huggingface.co/mlx-community/gemma-4-12B-coder-fable5-composer2.5-v1-OptiQ-4bit 
Base Model(s)  yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1   yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1
Model Size12b
Required VRAM8.8 GB
Updated2026-08-08
Maintainermlx-community
Model Typegemma4_unified
Model Files  5.3 GB: 1-of-2   3.5 GB: 2-of-2
Quantization Type4bit
Generates CodeYes
Model ArchitectureGemma4UnifiedForConditionalGeneration
Licenseapache-2.0
Transformers Version5.10.2
Tokenizer ClassGemmaTokenizer
Padding Token<pad>

Best Alternatives to Gemma 4 12B Coder Fable5 Composer2.5 V1 OptiQ 4bit

Best Alternatives
Context / RAM
Downloads
Likes
...le5 Composer2.5 V1 Abliterated0K / 24 GB300031
...r Fable5 Composer2.5 V1 Sft V50K / 23.9 GB300
...le5 Composer2.5 V1 Sft V5 Grpo0K / 23.9 GB240
...r Fable5 Composer2.5 V1 Sft V40K / 23.9 GB60
...poser2.5 V1 Sft V5 Abliterated0K / 24 GB1200
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/gemma-4-12B-coder-fable5-composer2.5-v1-OptiQ-4bit.