Smaugv0.1 4.0bpw H6 EXL2 by LoneStriker

 ยป  All LLMs  ยป  LoneStriker  ยป  Smaugv0.1 4.0bpw H6 EXL2   URL Share it on

  Merged Model   Autotrain compatible Base model:mistralai/mistral-7...   Conversational   Endpoints compatible   Exl2   License:apache-2.0   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow

Smaugv0.1 4.0bpw H6 EXL2 Benchmarks

Rank the Smaugv0.1 4.0bpw H6 EXL2 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  
Smaugv0.1 4.0bpw H6 EXL2 (LoneStriker/Smaugv0.1-4.0bpw-h6-exl2)

Best Alternatives to Smaugv0.1 4.0bpw H6 EXL2

Best Alternatives
HF Rank
Titanbagel64.2132K / 14.4 GB38820
...ernlm2 7B Llama 3.0bpw H6 EXL258.532K / 3.7 GB60
...ernlm2 7B Llama 4.0bpw H6 EXL258.532K / 4.5 GB70
...ernlm2 7B Llama 5.0bpw H6 EXL258.532K / 5.4 GB70
...ernlm2 7B Llama 6.0bpw H6 EXL258.532K / 6.3 GB80
...rnlm2 20B Llama 4.0bpw H6 EXL258.532K / 11 GB41
Vicuna 7B V1.555.274K / 13.5 GB689999241
Smaugv0.1 3.0bpw H6 EXL254.6195K / 13.9 GB31
Smaugv0.1 4.65bpw H6 EXL254.6195K / 20.8 GB31
Smaugv0.1 5.0bpw H6 EXL254.6195K / 22.3 GB33
Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/Smaugv0.1-4.0bpw-h6-exl2.

Smaugv0.1 4.0bpw H6 EXL2 Parameters and Internals

LLM NameSmaugv0.1 4.0bpw H6 EXL2
RepositoryOpen on ๐Ÿค— 
Base Model(s)  Mistral 7B V0.1   mistralai/Mistral-7B-v0.1
Merged ModelYes
Model Size7b
Required VRAM18 GB
Model Typellama
Model Files  8.6 GB: 1-of-3   8.5 GB: 2-of-3   0.9 GB: 3-of-3
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Context Length200000
Model Max Length200000
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size64000
Initializer Range0.02
Torch Data Typebfloat16

What open-source LLMs or SLMs are you in search of? 35042 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024042801