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

MiniMA 2 3B by GeneZC

 »  All LLMs  »  GeneZC  »  MiniMA 2 3B   URL Share it on

  837   

MiniMA 2 3B is an open-source language model by GeneZC. Features: 3b LLM, VRAM: 6GB, Context: 4K, License: apache-2.0, HF Score: 44.8, LLM Explorer Score: 0.28, ELO: 1190, Arc: 44.7, HellaSwag: 69.3, MMLU: 41.2, TruthfulQA: 38.4, WinoGrande: 66.7, GSM8K: 8.1.

  Arxiv:2311.07052   Dataset:eleutherai/pile   Dataset:p208p2002/wudao Dataset:togethercomputer/redpa...   En   Endpoints compatible   Llama   Model-index   Pytorch   Region:us   Safetensors   Zh
Model Card on HF 🤗: https://huggingface.co/GeneZC/MiniMA-2-3B 

MiniMA 2 3B Benchmarks

LMSys ELO: 1190 vs 1272 (so35)-6.4%
ARC: 44.71 vs 96.7 (so35)-53.8%
HellaSwag: 69.33 vs 95.3 (gpt4)-27.3%
MMLU: 41.22 vs 88.3 (so35)-53.3%
TruthfulQA: 38.44 vs 59 (gpt4)-34.8%
WinoGrande: 66.69 vs 87.5 (gpt4)-23.8%
GSM8K: 8.11 vs 96.4 (so35)-91.6%
LLME Score: 0.27519
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MiniMA 2 3B (GeneZC/MiniMA-2-3B)
🌟 Advertise your project 🚀

MiniMA 2 3B Parameters and Internals

Model Type 
text-generation
Additional Notes 
Derived from LLaMA-2, enhancements include better data mixture and increased number of trained tokens.
Supported Languages 
en (unknown), zh (unknown)
Training Details 
Data Sources:
EleutherAI/pile, togethercomputer/RedPajama-Data-1T, p208p2002/wudao
Methodology:
continued without distillation, better data mixture, more trained tokens
LLM NameMiniMA 2 3B
Repository 🤗https://huggingface.co/GeneZC/MiniMA-2-3B 
Model Size3b
Required VRAM6 GB
Updated2026-03-29
MaintainerGeneZC
Model Typellama
Model Files  6.0 GB   6.0 GB
Supported Languagesen zh
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.33.2
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size49216
Torch Data Typefloat16

Best Alternatives to MiniMA 2 3B

Best Alternatives
Context / RAM
Downloads
Likes
Nanbeige4.1 3B256K / 7.9 GB6765721019
Nanbeige4.1 3B Heretic256K / 7.9 GB61647
Nanbeige4.1 3B Bf16256K / 7.9 GB2643
Nanbeige4.1 3B Heretic256K / 7.9 GB60
ISA 03 Mini 3B Hybrid Preview256K / 6.5 GB8514
Llama 3.2 3B Instruct128K / 6.5 GB16749181724
Llama 3.2 3B128K / 6.5 GB1266692728
Llama 3.2 3B Instruct128K / 6.5 GB14269
Hermes 3 Llama 3.2 3B128K / 6.5 GB6378176
Llama 3.2 3B RP Toxic Fuse128K / 6.4 GB42
Note: green Score (e.g. "73.2") means that the model is better than GeneZC/MiniMA-2-3B.

Rank the MiniMA 2 3B 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