LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2 by Zoyd

 »  All LLMs  »  Zoyd  »  Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2   URL Share it on

  6   
  Merged Model   Arxiv:2402.13228   3-bit   Autotrain compatible   Conversational   Dataset:aqua rat Dataset:m-a-p/codefeedback-fil... Dataset:microsoft/orca-math-wo...   Endpoints compatible   Exl2   Instruct   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Zoyd/abacusai_Smaug-Llama-3-70B-Instruct-3_0bpw_exl2 

Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2 Benchmarks

LLME Score: 0.1865
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2 (Zoyd/abacusai_Smaug-Llama-3-70B-Instruct-3_0bpw_exl2)

Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2 Parameters and Internals

Model Type 
text generation
Use Cases 
Areas:
research, commercial applications
Applications:
text generation, multi-turn conversations
Primary Use Cases:
improving performance on real world multi-turn conversations
Additional Notes 
This version of Smaug uses new techniques and new data compared to previous versions.
Training Details 
Data Sources:
aqua_rat, microsoft/orca-math-word-problems-200k, m-a-p/CodeFeedback-Filtered-Instruction
Methodology:
Developed with new Smaug recipe for improving conversations
LLM NameAbacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2
Repository 🤗https://huggingface.co/Zoyd/abacusai_Smaug-Llama-3-70B-Instruct-3_0bpw_exl2 
Merged ModelYes
Model Size70b
Required VRAM28.6 GB
Updated2025-02-22
MaintainerZoyd
Model Typellama
Instruction-BasedYes
Model Files  8.5 GB: 1-of-4   8.5 GB: 2-of-4   8.6 GB: 3-of-4   3.0 GB: 4-of-4
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length8192
Model Max Length8192
Transformers Version4.40.0
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|end_of_text|>
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB75
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB63
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB52
...0B Instruct Gradient 262K 4bit256K / 39.7 GB93
...0B Instruct Gradient 262K 8bit256K / 75 GB72
...0B Instruct Gradient 262K 2bit256K / 21.9 GB51
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB50
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB81
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB70
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB60
Note: green Score (e.g. "73.2") means that the model is better than Zoyd/abacusai_Smaug-Llama-3-70B-Instruct-3_0bpw_exl2.

Rank the Abacusai Smaug Llama 3 70B Instruct 3 0bpw EXL2 Capabilities

🆘 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! 🌟

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? 43470 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer