Phi 4 Base by Sorawiz

 ยป  All LLMs  ยป  Sorawiz  ยป  Phi 4 Base   URL Share it on

  Merged Model   Arxiv:2306.01708   Autotrain compatible Base model:ngxson/lora-phi-4-a... Base model:prithivmlmods/phi-4... Base model:prithivmlmods/phi-4... Base model:prithivmlmods/phi-4... Base model:prithivmlmods/phi-4...   Conversational   Endpoints compatible   Llama   Lora   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF ๐Ÿค—: https://huggingface.co/Sorawiz/Phi-4-Base 

Phi 4 Base Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Phi 4 Base (Sorawiz/Phi-4-Base)

Phi 4 Base Parameters and Internals

LLM NamePhi 4 Base
Repository ๐Ÿค—https://huggingface.co/Sorawiz/Phi-4-Base 
Base Model(s)  Phi 4 Math IO   LoRA Phi 4 Abliterated   Phi 4 O1   LoRA Phi 4 Abliterated   Phi 4 QwQ   LoRA Phi 4 Abliterated   Phi 4 Empathetic   LoRA Phi 4 Abliterated   prithivMLmods/Phi-4-Math-IO   ngxson/LoRA-phi-4-abliterated   prithivMLmods/Phi-4-o1   ngxson/LoRA-phi-4-abliterated   prithivMLmods/Phi-4-QwQ   ngxson/LoRA-phi-4-abliterated   prithivMLmods/Phi-4-Empathetic   ngxson/LoRA-phi-4-abliterated
Merged ModelYes
Model Size14.7b
Required VRAM29.5 GB
Updated2025-02-05
MaintainerSorawiz
Model Typellama
Model Files  5.0 GB: 1-of-6   5.0 GB: 2-of-6   5.0 GB: 3-of-6   4.9 GB: 4-of-6   5.0 GB: 5-of-6   4.6 GB: 6-of-6
Model ArchitectureLlamaForCausalLM
Context Length16384
Model Max Length16384
Transformers Version4.46.2
Tokenizer ClassGPT2Tokenizer
Padding Token<|dummy_87|>
Vocabulary Size100352
LoRA ModelYes
Torch Data Typefloat16

Best Alternatives to Phi 4 Base

Best Alternatives
Context / RAM
Downloads
Likes
Phi 416K / 29.4 GB2215568
Phi 4 Model Stock V416K / 29.5 GB2457
Orca Mini Phi 416K / 29.4 GB3647
Phi 4 RR Shoup16K / 29.5 GB201
Phi 4 O116K / 29.4 GB65926
Phi 4 Model Stock16K / 29.5 GB2096
Phi 4 QwQ16K / 29.4 GB118713
Phi 4 RStock V0.116K / 29.5 GB140
Phi 4 Stock Ex16K / 29.5 GB382
Phi 4 Stock RP16K / 29.5 GB531
Note: green Score (e.g. "73.2") means that the model is better than Sorawiz/Phi-4-Base.

Rank the Phi 4 Base Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 42577 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227