LLM EXPLORER 60,645 MODELS INDEXED

Phi 3 Mini 4K Instruct Q4 by bongodongo

By bongodongo · 6 downloads
HF Repo ↗ Copy URL

Phi 3 Mini 4K Instruct Q4 is an open-source language model by bongodongo. Features: 3.9b LLM, VRAM: 2.3GB, Context: 4K, Quantized, Instruction-Based, LLM Explorer Score: 0.12.

  4-bit   Bitsandbytes   Conversational   Custom code   Endpoints compatible   Instruct   Phi3   Q4   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/bongodongo/phi-3-mini-4k-instruct-q4 

Phi 3 Mini 4K Instruct Q4 Parameters and Internals

Model Type 
quantized, causal language model
Additional Notes 
This version utilizes a 4-bit quantized model configuration using BitsAndBytesConfig with features such as double quantization and nf4 quant type for reduced computation requirements.
LLM NamePhi 3 Mini 4K Instruct Q4
Repository 🤗https://huggingface.co/bongodongo/phi-3-mini-4k-instruct-q4 
Base Model(s)  CitrusBoy/FinetunedModelV2.0   CitrusBoy/FinetunedModelV2.0
Model Size3.9b
Required VRAM2.3 GB
Updated2026-07-30
Maintainerbongodongo
Model Typephi3
Instruction-BasedYes
Model Files  2.3 GB
Quantization Typeq4
Model ArchitecturePhi3ForCausalLM
Context Length4096
Model Max Length4096
Transformers Version4.40.0
Tokenizer ClassLlamaTokenizer
Padding Token<|endoftext|>
Vocabulary Size32064
Torch Data Typefloat16

Best Alternatives to Phi 3 Mini 4K Instruct Q4

Best Alternatives
Context / RAM
Downloads
Likes
...Mini Instruct Unsloth Bnb 4bit128K / 3.2 GB678022
Phi 4 Mini Instruct Bnb 4bit128K / 2.9 GB68666
Phi 4 Mini Instruct Bnb 4bit128K / 2.9 GB50
...hi 3 Medium 128K Instruct 8bit128K / 14.9 GB182
...Mini 128K Instruct 4bit 64rank128K / 2.4 GB50
... 3 Mini 128K Instruct Bnb 4bit128K / 2.3 GB71
Phi 3 Medium 4K Instruct 8bit4K / 14.9 GB80
...3 Mini 4K Instruct 4bit 64rank4K / 2.4 GB70
...hi 3 Mini 4K Instruct Bnb 4bit4K / 2.3 GB196
...hi 3 Mini 4K Instruct Bnb 4bit4K / 2.3 GB70
Note: green Score (e.g. "73.2") means that the model is better than bongodongo/phi-3-mini-4k-instruct-q4.