Meta Llama Meta Llama 3 8B HQQ 1bit Smashed by PrunaAI

 ยป  All LLMs  ยป  PrunaAI  ยป  Meta Llama Meta Llama 3 8B HQQ 1bit Smashed   URL Share it on

  1bit   Autotrain compatible Base model:finetune:meta-llama... Base model:meta-llama/meta-lla...   Endpoints compatible   Llama   Pruna-ai   Quantized   Region:us

Meta Llama Meta Llama 3 8B HQQ 1bit Smashed Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
๐ŸŒŸ Advertise your project ๐Ÿš€

Meta Llama Meta Llama 3 8B HQQ 1bit Smashed Parameters and Internals

Model Type 
compression
Additional Notes 
Efficiency results were obtained after a hardware warmup. Smashed model may benefit specific use-cases under different configurations.
Training Details 
Data Sources:
WikiText
Methodology:
Compression with hqq
Input Output 
Input Format:
Text with tokens
Accepted Modalities:
text
Output Format:
Generated text
Performance Tips:
We recommend testing the efficiency gains directly in your use-cases.
LLM NameMeta Llama Meta Llama 3 8B HQQ 1bit Smashed
Repository ๐Ÿค—https://huggingface.co/PrunaAI/meta-llama-Meta-Llama-3-8B-HQQ-1bit-smashed 
Base Model(s)  Meta Llama 3 8B   meta-llama/Meta-Llama-3-8B
Model Size8b
Required VRAM3.1 GB
Updated2025-06-09
MaintainerPrunaAI
Model Typellama
Model Files  3.1 GB
Quantization Type1bit
Model ArchitectureLlamaForCausalLM
Context Length8192
Model Max Length8192
Transformers Version4.40.0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typefloat16
Meta Llama Meta Llama 3 8B HQQ 1bit Smashed (PrunaAI/meta-llama-Meta-Llama-3-8B-HQQ-1bit-smashed)

Best Alternatives to Meta Llama Meta Llama 3 8B HQQ 1bit Smashed

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 8.6 GB451
...B Instruct Gradient 1048K 4bit1024K / 4.5 GB102
...truct Gradient 1048K Bpw5 EXL21024K / 5.8 GB220
...truct Gradient 1048K Bpw6 EXL21024K / 6.7 GB132
Llama 3 8B Instruct 1048K 4bit1024K / 4.5 GB14225
Llama 3 8B Instruct 1048K 8bit1024K / 8.6 GB9917
...ct Gradient 1048K Bpw2.25 EXL21024K / 3.4 GB401
... Gradient 1048K 8.0bpw H8 EXL21024K / 8.6 GB113
Llama 3 8B Instruct 262K 2bit256K / 2.5 GB251
...B Instruct 262k V2 EXL2 8.0bpw256K / 8.4 GB220
Note: green Score (e.g. "73.2") means that the model is better than PrunaAI/meta-llama-Meta-Llama-3-8B-HQQ-1bit-smashed.

Rank the Meta Llama Meta Llama 3 8B HQQ 1bit Smashed Capabilities

๐Ÿ†˜ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐ŸŒŸ

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 48023 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124