LLM EXPLORER 61,918 MODELS INDEXED

ClipTagger 12B by inference-net

By inference-net · 20 downloads
HF Repo ↗ Copy URL

ClipTagger 12B is an open-source language model by inference-net. Features: 12b LLM, VRAM: 13.6GB, License: apache-2.0, LLM Explorer Score: 0.2.

  Compressed-tensors   Conversational   En   Gemma   Gemma3   Image-captioning   Image-text-to-text   Json-mode   Model-index   Region:us   Safetensors   Sharded   Structured-output   Tensorflow   Video-analysis   Video-understanding   Vlm
Model Card on HF 🤗: https://huggingface.co/inference-net/ClipTagger-12b 

ClipTagger 12B Parameters and Internals

LLM NameClipTagger 12B
Repository 🤗https://huggingface.co/inference-net/ClipTagger-12b 
Base Model(s)  google/gemma-12b   google/gemma-12b
Model Size12b
Required VRAM13.6 GB
Updated2026-08-08
Maintainerinference-net
Model Typegemma3
Model Files  5.0 GB: 1-of-3   5.0 GB: 2-of-3   3.6 GB: 3-of-3
Supported Languagesen
Model ArchitectureGemma3ForConditionalGeneration
Licenseapache-2.0
Transformers Version4.53.2
Tokenizer ClassGemmaTokenizer
Padding Token<pad>
Torch Data Typebfloat16

Best Alternatives to ClipTagger 12B

Best Alternatives
Context / RAM
Downloads
Likes
Fallen Gemma3 12B V1128K / 24.3 GB1824
Gemma 3 12B It0K / 24.3 GB2853479723
Gemma 3 12B It0K / 24.3 GB2582317
Translategemma 12B It0K / 24.3 GB12161324
Gemma 3 12B It Heretic V20K / 24.3 GB1132788
Gemma 3 12B It Heretic0K / 24.3 GB3358492
Gemma 3 12B Pt0K / 24.3 GB3684291
Gemma 3 12B It Law Fine Tuned0K / 24.3 GB740
Gemma 3 12B It FP8 Dynamic0K / 13.6 GB823
Gemma 3 12B Pt0K / 24.3 GB61065
Note: green Score (e.g. "73.2") means that the model is better than inference-net/ClipTagger-12b.