LLM EXPLORER 59,420 MODELS INDEXED

Phi 3 Mini 4K Instruct Bnb 4bit Ita by FairMind

By FairMind · 1181 downloads

Phi 3 Mini 4K Instruct Bnb 4bit Ita is an open-source language model by FairMind. Features: LLM, VRAM: 7.6GB, Context: 4K, License: cc-by-nc-sa-4.0, Quantized, Instruction-Based, LLM Explorer Score: 0.15, Arc: 63.3, HellaSwag: 78, MMLU: 67.4, GSM8K: 65.1.

  4bit   Autotrain compatible Base model:finetune:unsloth/ph... Base model:unsloth/phi-3-mini-...   Conversational Dataset:mchl-labs/stambecco da...   Endpoints compatible   Instruct   It   Italian   Mistral   Phi-3   Phi-3-mini   Pytorch   Quantized   Region:us   Sft   Sharded   Trl   Unsloth

Phi 3 Mini 4K Instruct Bnb 4bit Ita Parameters and Internals

Model Type 
text-generation-inference, transformers, trl, sft, phi-3, phi-3-mini, italian
Additional Notes 
This model was trained 2x faster with Unsloth and Huggingface's TRL library.
Supported Languages 
it (high)
LLM NamePhi 3 Mini 4K Instruct Bnb 4bit Ita
Repository πŸ€—https://huggingface.co/FairMind/Phi-3-mini-4k-instruct-bnb-4bit-Ita 
Base Model(s)  ...hi 3 Mini 4K Instruct Bnb 4bit   unsloth/Phi-3-mini-4k-instruct-bnb-4bit
Required VRAM7.6 GB
Updated2025-08-11
MaintainerFairMind
Model Typemistral
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   2.6 GB: 2-of-2
Supported Languagesit
Quantization Type4bit
Model ArchitectureMistralForCausalLM
Licensecc-by-nc-sa-4.0
Context Length4096
Model Max Length4096
Transformers Version4.40.1
Tokenizer ClassLlamaTokenizer
Padding Token<|endoftext|>
Vocabulary Size32064
Torch Data Typefloat16

Best Alternatives to Phi 3 Mini 4K Instruct Bnb 4bit Ita

Best Alternatives
Context / RAM
Downloads
Likes
...eZephir Sft Instruct Ead 16bit32K / 14.4 GB560
... Instruct Unsloth Merged 16bit4K / 7.6 GB61
... Mini 4K Instruct Bnb 4bit Ita4K / 7.6 GB15270
...ini New Model With Lora Merged4K / 7.6 GB160
Phired4K / 7.6 GB120
Phi 3 Mini Hospital Topic 504K / 7.6 GB340
Phi3 History V24K / 7.6 GB230
Model4K / 7.6 GB50
MainPHI34K / 7.6 GB90
Model4K / 7.6 GB70
Note: green Score (e.g. "73.2") means that the model is better than FairMind/Phi-3-mini-4k-instruct-bnb-4bit-Ita.