LLM EXPLORER 61,491 MODELS INDEXED

NeuralDaredevil 8B Abliterated by mlabonne

By mlabonne · 17515 downloads
HF Repo ↗ Copy URL

NeuralDaredevil 8B Abliterated is an open-source language model by mlabonne. Features: 8b LLM, VRAM: 16.1GB, Context: 8K, License: llama3, LLM Explorer Score: 0.24.

Base model:finetune:mlabonne/d... Base model:mlabonne/daredevil-...   Conversational Dataset:mlabonne/orpo-dpo-mix-...   Dpo   Endpoints compatible   Llama   Model-index   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated 

NeuralDaredevil 8B Abliterated Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
research, commercial applications
Applications:
role-playing
Additional Notes 
Model supports different quantization formats via various contributors.
Training Details 
Data Sources:
mlabonne/orpo-dpo-mix-40k
Methodology:
DPO fine-tuning
Release Notes 
Notes:
This model is a DPO fine-tune of mlabonne/Daredevil-8B-abliterated, trained on mlabonne/orpo-dpo-mix-40k dataset, improving performance by addressing performance loss due to prior abliteration.
LLM NameNeuralDaredevil 8B Abliterated
Repository 🤗https://huggingface.co/mlabonne/NeuralDaredevil-8B-abliterated 
Base Model(s)  Daredevil 8B Abliterated   mlabonne/Daredevil-8B-abliterated
Model Size8b
Required VRAM16.1 GB
Updated2026-08-03
Maintainermlabonne
Model Typellama
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.2
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typefloat16

Quantized Models of the NeuralDaredevil 8B Abliterated

Model
Likes
Downloads
VRAM
... Analysis June 05 2024 1 Epoch07716 GB
...alDaredevil 8B Abliterated AWQ01035 GB

Best Alternatives to NeuralDaredevil 8B Abliterated

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB1135125
UltraLong Thinking4192K / 16.1 GB23
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB12418
Cthulhu 8B V1.41048K / 16.1 GB1010
...raLong 1M Instruct Abliterated1048K / 32.1 GB49
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB70259
...xis Bookwriter Llama3.1 8B Sft1048K / 16.1 GB254