LLM EXPLORER 58,133 MODELS INDEXED

Phi 3 Mini 4K Instruct Onnx by microsoft

By microsoft · 200 downloads
HF Repo ↗ Copy URL

Phi 3 Mini 4K Instruct Onnx is an open-source language model by microsoft. Features: LLM, Context: 4K, License: mit, Instruction-Based, LLM Explorer Score: 0.13.

  Arxiv:2306.00978   Conversational   Custom code   Dml   Instruct   Onnx   Onnxruntime   Phi3   Region:us
Model Card on HF 🤗: https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx 

Phi 3 Mini 4K Instruct Onnx Parameters and Internals

Model Type 
ONNX
Additional Notes 
Phi-3 Mini is optimized for ONNX Runtime to accelerate inference on CPU and GPU and tested across diverse platforms.
Supported Languages 
Python (full), C (full), C++ (full)
Training Details 
Data Sources:
synthetic data, filtered websites
Methodology:
Supervised fine-tuning and direct preference optimization
Context Length:
4096
Hardware Used:
RTX 4090 (DirectML), 1 A100 80GB GPU, Standard_ND96amsr_A100_v4, Standard F64s v2, Samsung Galaxy S21
LLM NamePhi 3 Mini 4K Instruct Onnx
Repository 🤗https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-onnx 
Updated2026-07-19
Maintainermicrosoft
Model Typephi3
Instruction-BasedYes
Model ArchitecturePhi3ForCausalLM
Licensemit
Context Length4096
Model Max Length4096
Transformers Version4.39.3
Vocabulary Size32064
Torch Data Typebfloat16

Best Alternatives to Phi 3 Mini 4K Instruct Onnx

Best Alternatives
Context / RAM
Downloads
Likes
Phi 4 Mini Instruct ONNX GQA128K /  GB476
Phi 3.5 Mini Instruct Onnx128K /  GB13839
Phi 3.5 Mini Instruct Onnx Web128K /  GB40715
Phi 3 Mini 128K Instruct Onnx128K /  GB86193
...i 3 Mini 128K Instruct Ov Int4128K / 2 GB80
...Medium 128K Instruct Onnx Cuda128K /  GB8224
... Medium 128K Instruct Onnx Cpu128K /  GB8414
...um 128K Instruct Onnx Directml128K /  GB896
...3 Mini 128K Instruct Asym Int4128K / 2.5 GB1220
Meeting Model Phi3 128k128K / 7.7 GB50
Note: green Score (e.g. "73.2") means that the model is better than microsoft/Phi-3-mini-4k-instruct-onnx.