LLM EXPLORER 57,252 MODELS INDEXED

Paligemma2 28B Mix 224 by google

By google · 41 downloads
HF Repo ↗ Copy URL

Paligemma2 28B Mix 224 is an open-source language model by google. Features: 28b LLM, VRAM: 55.6GB, License: gemma, LLM Explorer Score: 0.15.

  Arxiv:1511.02283   Arxiv:1511.09207   Arxiv:1608.00272   Arxiv:1706.03762   Arxiv:1810.12440   Arxiv:1904.03493   Arxiv:1905.13648   Arxiv:1906.02467   Arxiv:1908.04913   Arxiv:2010.04295   Arxiv:2010.11929   Arxiv:2103.01913   Arxiv:2104.12756   Arxiv:2108.03353   Arxiv:2110.11624   Arxiv:2203.10244   Arxiv:2205.12522   Arxiv:2209.04372   Arxiv:2209.06794   Arxiv:2303.15343   Arxiv:2310.09199   Arxiv:2407.07726   Arxiv:2408.00118   Arxiv:2412.03555   Endpoints compatible   Image-text-to-text   Paligemma   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/google/paligemma2-28b-mix-224 

Paligemma2 28B Mix 224 Parameters and Internals

LLM NamePaligemma2 28B Mix 224
Repository 🤗https://huggingface.co/google/paligemma2-28b-mix-224 
Model Size28b
Required VRAM55.6 GB
Updated2026-08-09
Maintainergoogle
Model Typepaligemma
Model Files  4.8 GB: 1-of-12   4.9 GB: 2-of-12   5.0 GB: 3-of-12   4.9 GB: 4-of-12   4.9 GB: 5-of-12   5.0 GB: 6-of-12   4.9 GB: 7-of-12   4.9 GB: 8-of-12   5.0 GB: 9-of-12   4.9 GB: 10-of-12   4.9 GB: 11-of-12   1.5 GB: 12-of-12
Model ArchitecturePaliGemmaForConditionalGeneration
Licensegemma
Transformers Version4.47.0.dev0
Tokenizer ClassGemmaTokenizer
Padding Token<pad>
Torch Data Typebfloat16

Best Alternatives to Paligemma2 28B Mix 224

Best Alternatives
Context / RAM
Downloads
Likes
Paligemma2 28B Pt 8960K / 55.6 GB16452
Paligemma2 28B Mix 4480K / 55.6 GB20329
Paligemma2 28B Pt 4480K / 55.6 GB8912
Paligemma2 28B Pt 2240K / 55.6 GB878
Note: green Score (e.g. "73.2") means that the model is better than google/paligemma2-28b-mix-224.