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

Calme 4x7B MoE V0.2 by MaziyarPanahi

 »  All LLMs  »  MaziyarPanahi  »  Calme 4x7B MoE V0.2   URL Share it on

  9763   
  7b   Autotrain compatible   Calme   Conversational   Generated from trainer   Mistral   Mixtral   Moe   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/MaziyarPanahi/Calme-4x7B-MoE-v0.2 

Calme 4x7B MoE V0.2 Benchmarks

MMLU Pro: 22.86
GPQA: 3.91
MUSR: 12.54
BBH: 31.4
IFEval: 42.94 vs 88 (so35)-51.2%
ARC: 72.35 vs 96.7 (so35)-25.2%
HellaSwag: 88.63 vs 95.3 (gpt4)-7%
MMLU: 64.64 vs 88.3 (so35)-26.8%
TruthfulQA: 72.84 vs 59 (gpt4)23.5%
WinoGrande: 83.03 vs 87.5 (gpt4)-5.1%
GSM8K: 71.04 vs 96.4 (so35)-26.3%
MATH Lvl 5: 7.4
LLME Score: 0.24319
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Calme 4x7B MoE V0.2 (MaziyarPanahi/Calme-4x7B-MoE-v0.2)
🌟 Advertise your project 🚀

Calme 4x7B MoE V0.2 Parameters and Internals

Model Type 
Mixture of Experts (MoE), text generation
Use Cases 
Areas:
text generation
Applications:
educational tools, academic research, commonsense reasoning, natural language applications, trustworthy information retrieval, math educational software, context understanding
Additional Notes 
Quantized versions available for use on computers without accelerated hardware.
Supported Languages 
English (high proficiency), French (high proficiency), Ukrainian (high proficiency)
Training Details 
Methodology:
Mixture of Experts, fine-tuning of four individidual Calme-7B models
Model Architecture:
Integrates four Calme-7B models with two experts per token
Input Output 
Accepted Modalities:
text
Output Format:
text
LLM NameCalme 4x7B MoE V0.2
Repository 🤗https://huggingface.co/MaziyarPanahi/Calme-4x7B-MoE-v0.2 
Model NameCalme-4x7B-MoE-v0.2
Model CreatorMaziyarPanahi
Model Size24.2b
Required VRAM48.3 GB
Updated2025-09-23
MaintainerMaziyarPanahi
Model Typemixtral
Model Files  9.9 GB: 1-of-5   9.9 GB: 2-of-5   10.0 GB: 3-of-5   10.0 GB: 4-of-5   8.5 GB: 5-of-5
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.37.2
Tokenizer ClassLlamaTokenizer
Padding Token<s>
Vocabulary Size32000
Torch Data Typebfloat16

Quantized Models of the Calme 4x7B MoE V0.2

Model
Likes
Downloads
VRAM
Calme 4x7B MoE V0.2 GGUF4658 GB

Best Alternatives to Calme 4x7B MoE V0.2

Best Alternatives
Context / RAM
Downloads
Likes
Dzakwan MoE 4x7b Beta32K / 48.4 GB90720
Beyonder 4x7B V332K / 48.3 GB906559
Mera Mix 4x7B32K / 48.3 GB972519
Calme 4x7B MoE V0.132K / 48.3 GB90672
MixtureofMerges MoE 4x7b V532K / 48.3 GB87261
MixtureofMerges MoE 4x7b V432K / 48.3 GB90684
CognitiveFusion2 4x7B BF1632K / 48.3 GB97193
Proto Athena 4x7B32K / 48.4 GB50
Proto Athena V0.2 4x7B32K / 48.4 GB60
NeuralStar FusionWriter 4x7b32K / 48.3 GB97
Note: green Score (e.g. "73.2") means that the model is better than MaziyarPanahi/Calme-4x7B-MoE-v0.2.

Rank the Calme 4x7B MoE V0.2 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? 51545 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