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

Llama3 8B Spaetzle V20 by cstr

 »  All LLMs  »  cstr  »  Llama3 8B Spaetzle V20   URL Share it on

  7   

Llama3 8B Spaetzle V20 is an open-source language model by cstr. Features: 8b LLM, VRAM: 16GB, Context: 8K, License: llama3, Merged, HF Score: 71.8, LLM Explorer Score: 0.14, Arc: 70.4, HellaSwag: 85.7, MMLU: 68.5, TruthfulQA: 61, WinoGrande: 78.4, GSM8K: 67.

  Merged Model Base model:finetune:nbeerbower... Base model:nbeerbower/llama-3-...   Conversational   De   En   Endpoints compatible   Llama Nbeerbower/llama-3-wissenschaf...   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/cstr/llama3-8b-spaetzle-v20 

Llama3 8B Spaetzle V20 Benchmarks

ARC: 70.39 vs 96.7 (so35)-27.2%
HellaSwag: 85.69 vs 95.3 (gpt4)-10.1%
MMLU: 68.52 vs 88.3 (so35)-22.4%
TruthfulQA: 60.98 vs 59 (gpt4)3.4%
WinoGrande: 78.37 vs 87.5 (gpt4)-10.4%
GSM8K: 67.02 vs 96.4 (so35)-30.5%
LLME Score: 0.14232
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Llama3 8B Spaetzle V20 (cstr/llama3-8b-spaetzle-v20)
🌟 Advertise your project 🚀

Llama3 8B Spaetzle V20 Parameters and Internals

Model Type 
merge, mergekit, lazymergekit
Supported Languages 
English (native), German (native)
Input Output 
Input Format:
JSON format for prompts with roles and content
Accepted Modalities:
text
Output Format:
Generated text based on input prompts
Performance Tips:
For optimal performance, use appropriate hardware and configure parameters like temperature and top_k according to task requirements.
LLM NameLlama3 8B Spaetzle V20
Repository 🤗https://huggingface.co/cstr/llama3-8b-spaetzle-v20 
Base Model(s)  Llama 3 Wissenschaft 8B V2   nbeerbower/llama-3-wissenschaft-8B-v2
Merged ModelYes
Model Size8b
Required VRAM16 GB
Updated2026-04-11
Maintainercstr
Model Typellama
Model Files  1.1 GB: 1-of-9   1.9 GB: 2-of-9   2.0 GB: 3-of-9   1.9 GB: 4-of-9   2.0 GB: 5-of-9   1.9 GB: 6-of-9   2.0 GB: 7-of-9   1.9 GB: 8-of-9   1.3 GB: 9-of-9
Supported Languagesen de
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.39.3
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typebfloat16

Best Alternatives to Llama3 8B Spaetzle V20

Best Alternatives
Context / RAM
Downloads
Likes
...otron 8B UltraLong 4M Instruct4192K / 32.1 GB89119
...a 3.1 8B UltraLong 4M Instruct4192K / 32.1 GB17624
UltraLong Thinking4192K / 16.1 GB93
...a 3.1 8B UltraLong 2M Instruct2096K / 32.1 GB8759
...otron 8B UltraLong 2M Instruct2096K / 32.1 GB4017
Cthulhu 8B V1.41048K / 16.1 GB139
...otron 8B UltraLong 1M Instruct1048K / 32.1 GB344055
...a 3.1 8B UltraLong 1M Instruct1048K / 32.1 GB138729
...xis Bookwriter Llama3.1 8B Sft1048K / 16.1 GB124
Zero Llama 3.1 8B Beta61048K / 16.1 GB51
Note: green Score (e.g. "73.2") means that the model is better than cstr/llama3-8b-spaetzle-v20.

Rank the Llama3 8B Spaetzle V20 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? 52721 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