LLM EXPLORER 63,636 MODELS INDEXED

Gemma 4 A4B 98e V6 Coder It by ManniX-ITA

By ManniX-ITA · 22 downloads
HF Repo ↗ Copy URL

Gemma 4 A4B 98e V6 Coder It is an open-source language model by ManniX-ITA. Features: 26b LLM, VRAM: 41GB, License: apache-2.0, Code Generating, LLM Explorer Score: 0.23.

  Arxiv:2501.19393 Base model:finetune:google/gem... Base model:google/gemma-4-26b-...   Code   Codegen   Expert-pruning   Gemma4   Moe   Omnimergekit   Region:us   Safetensors   Sharded   Surgery   Tensorflow
Model Card on HF 🤗: https://huggingface.co/ManniX-ITA/gemma-4-A4B-98e-v6-coder-it 

Gemma 4 A4B 98e V6 Coder It Parameters and Internals

LLM NameGemma 4 A4B 98e V6 Coder It
Repository 🤗https://huggingface.co/ManniX-ITA/gemma-4-A4B-98e-v6-coder-it 
Base Model(s)  Gemma 4 26B A4B It   google/gemma-4-26B-A4B-it
Model Size26b
Required VRAM41 GB
Updated2026-08-07
MaintainerManniX-ITA
Model Typegemma4
Model Files  5.3 GB: 1-of-9   5.1 GB: 2-of-9   5.1 GB: 3-of-9   5.1 GB: 4-of-9   4.7 GB: 5-of-9   4.7 GB: 6-of-9   5.1 GB: 7-of-9   5.0 GB: 8-of-9   0.9 GB: 9-of-9
Generates CodeYes
Model ArchitectureGemma4ForConditionalGeneration
Licenseapache-2.0
Transformers Version5.5.0.dev0
Tokenizer ClassGemmaTokenizer
Padding Token<pad>