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

Meta Llama 3 8B Instruct by NousResearch

 »  All LLMs  »  NousResearch  »  Meta Llama 3 8B Instruct   URL Share it on

  123918   

Meta Llama 3 8B Instruct is an open-source language model by NousResearch. Features: 8b LLM, VRAM: 16.1GB, Context: 8K, License: other, Instruction-Based, LLM Explorer Score: 0.21, Arc: 61.4, HellaSwag: 78.6, MMLU: 67.1, GSM8K: 69.4.

  Conversational   Deploy:azure   En   Endpoints compatible   Facebook   Instruct   Llama   Llama-3   Meta   Pytorch   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct 

Meta Llama 3 8B Instruct Benchmarks

ARC: 61.43 vs 96.7 (so35)-36.5%
HellaSwag: 78.58 vs 95.3 (gpt4)-17.5%
MMLU: 67.08 vs 88.3 (so35)-24%
GSM8K: 69.37 vs 96.4 (so35)-28%
LLME Score: 0.21414
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Meta Llama 3 8B Instruct (NousResearch/Meta-Llama-3-8B-Instruct)
🌟 Advertise your project 🚀

Meta Llama 3 8B Instruct Parameters and Internals

Model Type 
text generation, instruction tuned
Use Cases 
Areas:
commercial, research
Applications:
instruction-tuned for chat applications
Primary Use Cases:
chat-oriented generative tasks
Limitations:
Not suitable for language other than English, restricted under Acceptable Use Policy
Considerations:
Developers encouraged to implement safety assessments for specific applications.
Additional Notes 
Model supports fine-tuning for languages beyond English under compliance with the license and use policy.
Supported Languages 
en (Native)
Training Details 
Data Sources:
publicly available online data
Data Volume:
15 trillion tokens
Methodology:
supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF)
Context Length:
8000
Training Time:
7.7M GPU hours on H100-80GB
Hardware Used:
H100-80GB GPUs
Model Architecture:
optimized transformer architecture
Safety Evaluation 
Methodologies:
red teaming, adversarial evaluations
Findings:
residual risks remain, model refusals reduced
Risk Categories:
CBRNE, cybersecurity, child safety
Ethical Considerations:
Ethical considerations include avoiding misuse of AI in harmful areas.
Responsible Ai Considerations 
Fairness:
Model is optimized to balance helpfulness and alignment, with considerations for avoiding biases.
Transparency:
Open source release with detailed documentation and responsible use guidelines.
Accountability:
Users must comply with license terms and acceptable use policy.
Mitigation Strategies:
Safety tools like Meta Llama Guard and Code Shield provided.
Input Output 
Input Format:
text input
Accepted Modalities:
text
Output Format:
text and code output
Performance Tips:
Use appropriate hardware and fine-tuning methods for optimal performance.
Release Notes 
Version:
April 18, 2024
Date:
2024-04-18
Notes:
Initial release of Llama 3 models.
LLM NameMeta Llama 3 8B Instruct
Repository 🤗https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct 
Model Size8b
Required VRAM16.1 GB
Updated2026-06-09
MaintainerNousResearch
Model Typellama
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   4.9 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length8192
Model Max Length8192
Transformers Version4.40.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Meta Llama 3 8B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB230121
UltraLong Thinking4192K / 16.1 GB53
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB8218
Cthulhu 8B V1.41048K / 16.1 GB129
...raLong 1M Instruct Abliterated1048K / 32.1 GB569
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB65259
Zero Llama 3.1 8B Beta61048K / 16.1 GB41
Note: green Score (e.g. "73.2") means that the model is better than NousResearch/Meta-Llama-3-8B-Instruct.

Rank the Meta Llama 3 8B Instruct 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? 54415 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer