LLM EXPLORER 62,404 MODELS INDEXED

Chargoddard Llama3 42B V0.4.0bpw EXL2 by lucyknada

By lucyknada · 4 downloads
HF Repo ↗ Copy URL

Chargoddard Llama3 42B V0.4.0bpw EXL2 is an open-source language model by lucyknada. Features: 42b LLM, VRAM: 23.4GB, Context: 8K, License: other, Quantized, LLM Explorer Score: 0.12.

  Arxiv:2403.17887   4-bit   Axolotl   Conversational   Dataset:jeankaddour/minipile   En   Endpoints compatible   Exl2   Llama   Mergekit   Pytorch   Quantized   Region:us   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/lucyknada/chargoddard_llama3-42b-v0-4.0bpw-EXL2 

Chargoddard Llama3 42B V0.4.0bpw EXL2 Parameters and Internals

Model Type 
text generation
Additional Notes 
The model has been pruned to 42B parameters from the original 70B using novel pruning methodology and requires further evaluation.
Supported Languages 
en (high)
Training Details 
Data Sources:
JeanKaddour/minipile
Data Volume:
~100M tokens
Methodology:
Pruned using The Unreasonable Ineffectiveness of the Deeper Layers and PruneMe. Post-pruning training using QLoRA.
LLM NameChargoddard Llama3 42B V0.4.0bpw EXL2
Repository 🤗https://huggingface.co/lucyknada/chargoddard_llama3-42b-v0-4.0bpw-EXL2 
Model Size42b
Required VRAM23.4 GB
Updated2026-08-01
Maintainerlucyknada
Model Typellama
Model Files  8.5 GB: 1-of-3   8.5 GB: 2-of-3   6.4 GB: 3-of-3
Supported Languagesen
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length8192
Model Max Length8192
Transformers Version4.40.0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|end_of_text|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Chargoddard Llama3 42B V0.4.0bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
Llama3 42B V08K / 86.8 GB12117
Llama 42M1K / 0.1 GB90
Tinyllama 42M1K / 0.1 GB251
Tinyllama 42M Fp321K / 0.2 GB63
Note: green Score (e.g. "73.2") means that the model is better than lucyknada/chargoddard_llama3-42b-v0-4.0bpw-EXL2.