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

MiniChat 1.5 3B by GeneZC

 »  All LLMs  »  GeneZC  »  MiniChat 1.5 3B   URL Share it on

  2159   
  Arxiv:2305.18290   Arxiv:2310.05914   Arxiv:2311.07052   Autotrain compatible   Conversational   En   Endpoints compatible   Llama   Model-index   Pytorch   Region:us   Safetensors   Zh
Model Card on HF 🤗: https://huggingface.co/GeneZC/MiniChat-1.5-3B 

MiniChat 1.5 3B Benchmarks

ARC: 46.5 vs 96.7 (so35)-51.9%
HellaSwag: 68.28 vs 95.3 (gpt4)-28.4%
MMLU: 46.67 vs 88.3 (so35)-47.1%
TruthfulQA: 50.71 vs 59 (gpt4)-14.1%
WinoGrande: 65.04 vs 87.5 (gpt4)-25.7%
GSM8K: 24.18 vs 96.4 (so35)-74.9%
LLME Score: 0.20167
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MiniChat 1.5 3B (GeneZC/MiniChat-1.5-3B)

MiniChat 1.5 3B Parameters and Internals

Model Type 
text-generation
Use Cases 
Areas:
research, commercial applications
Additional Notes 
Derived from LLaMA2; Outperforming various 3B competitors.
Supported Languages 
en (English), zh (Chinese)
Training Details 
Methodology:
Fine-tuned from an adapted version of LLaMA2-7B following "Towards the Law of Capacity Gap in Distilling Language Models". Better data mixture; use of NEFTune; use of DPO.
Input Output 
Accepted Modalities:
text
Performance Tips:
Use multi-turn conversation by continuously appending questions to the `conv`. Set `torch_dtype` to `torch.float16` for better performance on compatible hardware.
Release Notes 
Version:
MiniChat-1.5-3B
Notes:
Better data mixture; use of NEFTune; use of DPO. Derived from LLaMA2.
LLM NameMiniChat 1.5 3B
Repository 🤗https://huggingface.co/GeneZC/MiniChat-1.5-3B 
Model Size3b
Required VRAM6 GB
Updated2025-02-22
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 MiniChat 1.5 3B

Best Alternatives
Context / RAM
Downloads
Likes
Llama 3.2 3B Instruct128K / 6.5 GB16030891085
Llama 3.2 3B128K / 6.5 GB292157509
Hermes 3 Llama 3.2 3B128K / 6.5 GB21968144
ReasoningCore 3B RE1 V2128K / 6.5 GB1160
...penReasoner Llama 3.2 3B Rs1.0128K / 6.5 GB1471
Zeitgeist 3B V1128K / 6.5 GB492
...eflection L3.2 JametMiniMix 3B128K / 6.4 GB1100
Calme 3.1 Llamaloi 3B128K / 10.6 GB29111
Dolphin3.0 Llama3.2 3B128K / 6.5 GB1880036
... 3.2 3B Math Instruct RE1 ORPO128K / 6.5 GB1350
Note: green Score (e.g. "73.2") means that the model is better than GeneZC/MiniChat-1.5-3B.

Rank the MiniChat 1.5 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? 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