Llama 3 8B Instruct GPTQ 4 Bit by astronomer

 ยป  All LLMs  ยป  astronomer  ยป  Llama 3 8B Instruct GPTQ 4 Bit   URL Share it on

  Arxiv:2210.17323   4-bit   Astronomer   Autotrain compatible Base model:meta-llama/meta-lla... Base model:quantized:meta-llam...   Conversational   Dataset:wikitext   Endpoints compatible   Facebook   Finetuned   Gptq   Instruct   Llama   Llama-3   Meta   Pretrained   Quantized   Region:us   Safetensors

Llama 3 8B Instruct GPTQ 4 Bit Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 8B Instruct GPTQ 4 Bit (astronomer/Llama-3-8B-Instruct-GPTQ-4-Bit)

Llama 3 8B Instruct GPTQ 4 Bit Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model is quantized to 4-bit using GPTQ with small accuracy degradation. It is optimized for serving with reduced VRAM usage.
Input Output 
Input Format:
<|begin_of_text|><|start_header_id|>user<|end_header_id|> {{prompt}}<|eot_id|> <|start_header_id|>assistant<|end_header_id|>
LLM NameLlama 3 8B Instruct GPTQ 4 Bit
Repository ๐Ÿค—https://huggingface.co/astronomer/Llama-3-8B-Instruct-GPTQ-4-Bit 
Model NameMeta-Llama-3-8B-Instruct
Model Creatorastronomer-io
Base Model(s)  Meta Llama 3 8B Instruct   meta-llama/Meta-Llama-3-8B-Instruct
Model Size8b
Required VRAM5.7 GB
Updated2024-12-23
Maintainerastronomer
Model Typellama
Instruction-BasedYes
Model Files  5.7 GB
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length8192
Model Max Length8192
Transformers Version4.38.2
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3 8B Instruct GPTQ 4 Bit

Best Alternatives
Context / RAM
Downloads
Likes
...a 3 8B Instruct 262K 4bit GPTQ256K / 5.8 GB91
... 8B Instruct 262K 4bit GPTQ 02256K / 5.7 GB70
...lama 3.1 8B Instruct GPTQ INT4128K / 5.8 GB8876021
...Instruct 80K Qlora Merged GPTQ80K / 5.8 GB210
...oLeo Instruct 8B 32K V0.1 GPTQ64K / 5.7 GB70
...eta Llama 3 8B Instruct Marlin8K / 5.7 GB40750
Llama 3 8B Instruct GPTQ 8 Bit8K / 9.7 GB430025
Meta Llama 3 8B Instruct GPTQ8K / 5.7 GB7213
Meta Llama 3 8B Instruct GPTQ8K / 5.8 GB100
Meta Llama 3 8B Instruct GPTQ8K / 5.8 GB141
Note: green Score (e.g. "73.2") means that the model is better than astronomer/Llama-3-8B-Instruct-GPTQ-4-Bit.

Rank the Llama 3 8B Instruct GPTQ 4 Bit 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? 40126 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217