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 8B Ultra Instruct by Dampfinchen

 »  All LLMs  »  Dampfinchen  »  Llama 3 8B Ultra Instruct   URL Share it on

  926   
  Arxiv:2306.01708   Arxiv:2311.03099   Autotrain compatible Base model:aaditya/llama3-open... Base model:jondurbin/bagel-8b-... Base model:merge:aaditya/llama... Base model:merge:jondurbin/bag... Base model:merge:undi95/llama-... Base model:merge:undi95/meta-l... Base model:merge:vagosolutions... Base model:merge:weyaxi/einste... Base model:undi95/llama-3-lewd... Base model:undi95/meta-llama-3... Base model:vagosolutions/llama... Base model:weyaxi/einstein-v6....   Conversational   Endpoints compatible   Instruct   Llama   Merge   Mergekit   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Dampfinchen/Llama-3-8B-Ultra-Instruct 

Llama 3 8B Ultra Instruct Benchmarks

LLME Score: 0.1758
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 Ultra Instruct (Dampfinchen/Llama-3-8B-Ultra-Instruct)
🌟 Advertise your project 🚀

Llama 3 8B Ultra Instruct Parameters and Internals

Model Type 
General Purpose, Roleplaying
Use Cases 
Areas:
Research, Roleplaying, Medical Knowledge
Applications:
Better RAG capabilities, General Intelligence, Vision support, Medical field knowledge
Primary Use Cases:
Enticing roleplaying, Multilingual German support
Additional Notes 
The model combines intelligent instruct capabilities with specialized roleplaying features. It is considered dense and does not need a lot more data beyond its final weight adjustment of 0.65 in the ties merge to preserve its core capabilities.
Training Details 
Data Sources:
Undi95, Jon Durbin, Aaditya, VAGOsolutions, Teknium, Camel
Methodology:
Merged using DARE and TIES. Final merge weights conservatively adjusted to 0.65 to preserve Llama Instruct's intelligence.
Hardware Used:
Elinas provided compute
Input Output 
Input Format:
<|begin_of_text|><|start_header_id|>system<|end_header_id|> {system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|> {input}<|eot_id|><|start_header_id|>assistant<|end_header_id|> {output}<|eot_id|>
Accepted Modalities:
Text
Output Format:
Text
LLM NameLlama 3 8B Ultra Instruct
Repository 🤗https://huggingface.co/Dampfinchen/Llama-3-8B-Ultra-Instruct 
Base Model(s)  Meta Llama 3 8B Instruct Hf   Llama 3 LewdPlay 8B Evo   jondurbin/bagel-8b-v1.0   Weyaxi/Einstein-v6.1-Llama3-8B   ...ama 3 SauerkrautLM 8B Instruct   aaditya/OpenBioLLM-Llama3-8B   Undi95/Meta-Llama-3-8B-Instruct-hf   Undi95/Llama-3-LewdPlay-8B-evo   jondurbin/bagel-8b-v1.0   Weyaxi/Einstein-v6.1-Llama3-8B   VAGOsolutions/Llama-3-SauerkrautLM-8b-Instruct   aaditya/OpenBioLLM-Llama3-8B
Model Size8b
Required VRAM16 GB
Updated2025-07-29
MaintainerDampfinchen
Model Typellama
Instruction-BasedYes
Model Files  9.9 GB: 1-of-2   6.1 GB: 2-of-2
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.0.dev0
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama 3 8B Ultra Instruct

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB47307115
UltraLong Thinking4192K / 16.1 GB5172
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB84415
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB149847
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
Zero Llama 3.1 8B Beta61048K / 16.1 GB4721
...dger Nu Llama 3.1 8B UltraLong1048K / 16.2 GB63
....1 1million Ctx Dark Planet 8B1048K / 32.3 GB102
Note: green Score (e.g. "73.2") means that the model is better than Dampfinchen/Llama-3-8B-Ultra-Instruct.

Rank the Llama 3 8B Ultra 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? 50230 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