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

Huh 1 by yweslakarep

 »  All LLMs  »  yweslakarep  »  Huh 1   URL Share it on

  58   
  Merged Model   Autotrain compatible   Barrahome/mistroll-7b-v2.2 Base model:barrahome/mistroll-... Base model:huggingfaceh4/zephy...   Conversational   Endpoints compatible   Huggingfaceh4/zephyr-7b-beta   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/yweslakarep/huh-1 

Huh 1 Benchmarks

ARC: 68.6 vs 96.7 (so35)-29.1%
HellaSwag: 86.8 vs 95.3 (gpt4)-8.9%
MMLU: 64.25 vs 88.3 (so35)-27.2%
TruthfulQA: 67.41 vs 59 (gpt4)14.3%
WinoGrande: 81.06 vs 87.5 (gpt4)-7.4%
GSM8K: 58.53 vs 96.4 (so35)-39.3%
LLME Score: 0.20015
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Huh 1 (yweslakarep/huh-1)

Huh 1 Parameters and Internals

Additional Notes 
huh-1 is a merge of HuggingFaceH4/zephyr-7b-beta and BarraHome/Mistroll-7B-v2.2 using LazyMergekit with a specific configuration.
LLM NameHuh 1
Repository 🤗https://huggingface.co/yweslakarep/huh-1 
Base Model(s)  Zephyr 7B Beta   BarraHome/Mistroll-7B-v2.2   HuggingFaceH4/zephyr-7b-beta   BarraHome/Mistroll-7B-v2.2
Merged ModelYes
Model Size7b
Required VRAM14.5 GB
Updated2025-02-22
Maintaineryweslakarep
Model Typemistral
Model Files  1.0 GB: 1-of-15   1.0 GB: 2-of-15   1.0 GB: 3-of-15   1.0 GB: 4-of-15   0.9 GB: 5-of-15   1.0 GB: 6-of-15   1.0 GB: 7-of-15   1.0 GB: 8-of-15   0.9 GB: 9-of-15   1.0 GB: 10-of-15   1.0 GB: 11-of-15   1.0 GB: 12-of-15   0.9 GB: 13-of-15   1.0 GB: 14-of-15   0.8 GB: 15-of-15
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.41.0
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Huh 1

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB462011
MegaBeam Mistral 7B 512K512K / 14.4 GB568150
SpydazWeb AI HumanAI RP512K / 14.4 GB121
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB563316
Hebrew Mistral 7B 200K256K / 30 GB1461915
Astral 256K 7B V2250K / 14.4 GB70
Astral 256K 7B250K / 14.4 GB50
Test001128K / 14.5 GB90
Note: green Score (e.g. "73.2") means that the model is better than yweslakarep/huh-1.

Rank the Huh 1 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