LLM EXPLORER 65,999 MODELS INDEXED

NVIDIA Nemotron 3.5 Bolt E4B BF16 Heretic by darkc0de

By darkc0de · 109 downloads
HF Repo ↗ Copy URL

NVIDIA Nemotron 3.5 Bolt E4B BF16 Heretic is an open-source language model by darkc0de. Features: 7.9b LLM, VRAM: 15.9GB, LLM Explorer Score: 0.3.

  Arxiv:1910.09700   Conversational   Endpoints compatible   Gemma4   Image-text-to-text   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/darkc0de/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16-heretic 

NVIDIA Nemotron 3.5 Bolt E4B BF16 Heretic Parameters and Internals

LLM NameNVIDIA Nemotron 3.5 Bolt E4B BF16 Heretic
Repository 🤗https://huggingface.co/darkc0de/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16-heretic 
Model Size7.9b
Required VRAM15.9 GB
Updated2026-10-28
Maintainerdarkc0de
Model Typegemma4
Model Files  5.6 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   0.3 GB: 4-of-4
Model ArchitectureGemma4ForConditionalGeneration
Transformers Version5.20.2
Tokenizer ClassGemmaTokenizer
Padding Token<pad>

Best Alternatives to NVIDIA Nemotron 3.5 Bolt E4B BF16 Heretic

Best Alternatives
Context / RAM
Downloads
Likes
Basal Omni 1.0 Mini0K / 15.9 GB10943
Baihua Gemma4 E4B0K / 15.9 GB1353
Kitsune Tales E4B EN0K / 15.9 GB2981
Kitsune Tales E4B JP0K / 15.9 GB2891
Firehouse Cactus 1.00K / 15.9 GB7360
ContextPilot E4B0K / 15.9 GB4045
Quatfit Mini0K / 31.8 GB10747
Gemma 4 E4B It NVFP4A160K / 10.3 GB311
Gemma 4 E4B It NVFP40K / 10.3 GB311
Gemma4 Merged K8s0K / 15.9 GB1480
Note: green Score (e.g. "73.2") means that the model is better than darkc0de/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16-heretic.