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

Lumosia V2 MoE 4x10.7 by Steelskull

 »  All LLMs  »  Steelskull  »  Lumosia V2 MoE 4x10.7   URL Share it on

  75   
  Autotrain compatible   Conversational   Endpoints compatible   Lumosia   Mixtral   Model-index   Moe   Region:us   Safetensors   Sharded   Solar   Solar moe   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Steelskull/Lumosia-v2-MoE-4x10.7 

Lumosia V2 MoE 4x10.7 Benchmarks

ARC: 70.39 vs 96.7 (so35)-27.2%
HellaSwag: 87.87 vs 95.3 (gpt4)-7.8%
MMLU: 66.45 vs 88.3 (so35)-24.7%
TruthfulQA: 68.48 vs 59 (gpt4)16.1%
WinoGrande: 84.21 vs 87.5 (gpt4)-3.8%
GSM8K: 65.13 vs 96.4 (so35)-32.4%
LLME Score: 0.1427
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Lumosia V2 MoE 4x10.7 (Steelskull/Lumosia-v2-MoE-4x10.7)
🌟 Advertise your project 🚀

Lumosia V2 MoE 4x10.7 Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model supports a context length of up to 16k and employs a mixture of experts architecture.
LLM NameLumosia V2 MoE 4x10.7
Repository 🤗https://huggingface.co/Steelskull/Lumosia-v2-MoE-4x10.7 
Model Size36.1b
Required VRAM72.3 GB
Updated2024-10-16
MaintainerSteelskull
Model Typemixtral
Model Files  9.9 GB: 1-of-8   10.0 GB: 2-of-8   10.0 GB: 3-of-8   10.0 GB: 4-of-8   10.0 GB: 5-of-8   10.0 GB: 6-of-8   10.0 GB: 7-of-8   2.4 GB: 8-of-8
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.37.1
Tokenizer ClassLlamaTokenizer
Padding Token<s>
Vocabulary Size32000
Torch Data Typebfloat16

Best Alternatives to Lumosia V2 MoE 4x10.7

Best Alternatives
Context / RAM
Downloads
Likes
Umbra V3 MoE 4x11b 2ex32K / 72.3 GB2864
PiVoT MoE32K / 72.3 GB17908
Umbra V3 MoE 4x11b 2ex32K / 72.3 GB54
Umbra V3 MoE 4x11b32K / 72.3 GB55
Umbra V2.1 MoE 4x10.732K / 72.3 GB66
Mixolar 4x7b4K / 72.3 GB97803
Smartsolmix 4x10.7B V14K / 72.3 GB18580
Orca SOLAR 4x10.7B4K / 72.3 GB17380
MetaModel MoE4K / 72.3 GB19140
SOLARC MoE 10.7Bx44K / 144.7 GB19179
Note: green Score (e.g. "73.2") means that the model is better than Steelskull/Lumosia-v2-MoE-4x10.7.

Rank the Lumosia V2 MoE 4x10.7 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? 51535 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