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

Llama 3 70B Instruct Gradient 1048K AWQ by starsy

 »  All LLMs  »  starsy  »  Llama 3 70B Instruct Gradient 1048K AWQ   URL Share it on

  13   
  Arxiv:2305.14233   Arxiv:2309.00071   Arxiv:2310.05209   Arxiv:2402.08268   4-bit   Autotrain compatible   Awq   Conversational   En   Endpoints compatible   Instruct   Llama   Llama-3   Meta   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/starsy/Llama-3-70B-Instruct-Gradient-1048k-AWQ 

Llama 3 70B Instruct Gradient 1048K AWQ Benchmarks

ARC: 66.81 vs 96.7 (so35)-30.9%
HellaSwag: 85.46 vs 95.3 (gpt4)-10.3%
MMLU: 76.37 vs 88.3 (so35)-13.5%
TruthfulQA: 53.73 vs 59 (gpt4)-8.9%
WinoGrande: 82.64 vs 87.5 (gpt4)-5.6%
GSM8K: 78.85 vs 96.4 (so35)-18.2%
LLME Score: 0.16105
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama 3 70B Instruct Gradient 1048K AWQ (starsy/Llama-3-70B-Instruct-Gradient-1048k-AWQ)
🌟 Advertise your project 🚀

Llama 3 70B Instruct Gradient 1048K AWQ Parameters and Internals

Model Type 
text generation, instruction-tuned
Use Cases 
Areas:
commercial applications, research
Applications:
assistant-like chat
Primary Use Cases:
Natural language generation tasks
Limitations:
Use only in accordance with laws
Considerations:
Safety testing and tuning recommended
Additional Notes 
Uses proprietary EasyContext Blockwise RingAttention library for long context training.
Supported Languages 
English (Native)
Training Details 
Data Sources:
SlimPajama, UltraChat, publicly available instruction datasets
Data Volume:
15 trillion tokens
Methodology:
NTK-aware interpolation
Context Length:
1048000
Training Time:
100-516 minutes per context length
Hardware Used:
NVIDIA L40S, H100-80GB
Model Architecture:
auto-regressive transformer architecture
Safety Evaluation 
Methodologies:
red-teaming, adversarial evaluations, CyberSecEval
Findings:
residual risks remain, over-refusal reduced
Risk Categories:
CBRNE, Cyber Security, Child Safety
Ethical Considerations:
Open approach for better, safer products
Responsible Ai Considerations 
Fairness:
Access to many different backgrounds
Transparency:
Open source approach
Accountability:
Developers responsible for safe use
Mitigation Strategies:
Meta Llama Guard and Code Shield safeguards
Input Output 
Input Format:
Text-only
Accepted Modalities:
text
Output Format:
Text and code
Performance Tips:
Not provided
Release Notes 
Version:
April 18, 2024
Date:
April 18, 2024
Notes:
Initial release of Llama 3 model
LLM NameLlama 3 70B Instruct Gradient 1048K AWQ
Repository 🤗https://huggingface.co/starsy/Llama-3-70B-Instruct-Gradient-1048k-AWQ 
Base Model(s)  ... 3 70B Instruct Gradient 1048K   gradientai/Llama-3-70B-Instruct-Gradient-1048k
Model Size70b
Required VRAM39.9 GB
Updated2025-09-26
Maintainerstarsy
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-9   4.9 GB: 2-of-9   4.9 GB: 3-of-9   4.9 GB: 4-of-9   4.9 GB: 5-of-9   4.9 GB: 6-of-9   4.9 GB: 7-of-9   3.4 GB: 8-of-9   2.1 GB: 9-of-9
Supported Languagesen
AWQ QuantizationYes
Quantization Typeawq
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length1048576
Model Max Length1048576
Transformers Version4.40.2
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typefloat16

Best Alternatives to Llama 3 70B Instruct Gradient 1048K AWQ

Best Alternatives
Context / RAM
Downloads
Likes
...70B Instruct Gradient 262K AWQ256K / 39.9 GB60
Llama 3.3 70B Instruct AWQ128K / 39.9 GB3935456
Llama 3.3 70B Instruct AWQ128K / 39.9 GB17963535
...lama 3.3 70B Instruct AWQ INT4128K / 39.9 GB4584025
... SauerkrautLM 70B Instruct AWQ128K / 39.9 GB475
Llama 3 70B Instruct AWQ8K / 39.9 GB216869
Meta Llama 3 70B Instruct AWQ8K / 39.9 GB80012
Meta Llama 3 70B Instruct AWQ8K / 39.9 GB81
Llama 3 70B Instruct AWQ8K / 39.9 GB51
... Llama 3 70B Instruct AWQ GEMM8K / 39.9 GB60
Note: green Score (e.g. "73.2") means that the model is better than starsy/Llama-3-70B-Instruct-Gradient-1048k-AWQ.

Rank the Llama 3 70B Instruct Gradient 1048K AWQ 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? 51544 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 v20241124
  Support LLM Explorer