Phi-3.5-mini-instruct-GGUF (Platinum Series)
This repository contains the Platinum Series universal GGUF release of Phi-3.5-mini-instruct. This collection provides multiple quantization levels optimized for cross-platform performance, offering advanced reasoning capabilities with 128k context support.
π¦ Available Files & Quantization Details
| File Name | Quantization | Size | Accuracy | Recommended For |
|---|---|---|---|---|
| Phi-3.5-mini-instruct-Platinum-F16.gguf | FP16 | ~7.6 GB | 100% | Master Reference / Benchmarking |
| Phi-3.5-mini-instruct-Platinum-Q8_0.gguf | Q8_0 | ~4.1 GB | 99.9% | Platinum Reference / High-Fidelity |
| Phi-3.5-mini-instruct-Platinum-Q6_K.gguf | Q6_K | ~3.1 GB | 99.8% | High-Quality Inference |
| Phi-3.5-mini-instruct-Platinum-Q5_K_M.gguf | Q5_K_M | ~2.8 GB | 99.4% | Balanced Desktop Performance |
| Phi-3.5-mini-instruct-Platinum-Q4_K_M.gguf | Q4_K_M | ~2.4 GB | 98.8% | Mobile / Low-Power Efficiency |
π Python Inference (llama-cpp-python)
To run these engines using Python:
from llama_cpp import Llama
llm = Llama(
model_path="Phi-3.5-mini-instruct-Platinum-Q8_0.gguf",
n_gpu_layers=-1, # Target all layers to NVIDIA/Apple GPU
n_ctx=4096
)
output = llm("Discuss the architectural benefits of Phi-3.5.", max_tokens=150)
print(output["choices"][0]["text"])
π» For C# / .NET Users (LLamaSharp)
This collection is fully compatible with .NET applications via the LLamaSharp library.
using LLama.Common;
using LLama;
var parameters = new ModelParams("Phi-3.5-mini-instruct-Platinum-Q8_0.gguf") {
ContextSize = 4096,
GpuLayerCount = 35
};
using var model = LLamaWeights.LoadFromFile(parameters);
using var context = model.CreateContext(parameters);
var executor = new InteractiveExecutor(context);
Console.WriteLine("Universal Engine Active.");
ποΈ Technical Details
- Optimization Tool: llama.cpp (CUDA-accelerated)
- Architecture: Phi-3.5
- Hardware Validation: Dual-GPU (RTX 3090 + RTX A4000)
β Support the Forge
Maintaining the production line for high-fidelity models requires significant hardware resources. If these tools power your research or industrial projects, please consider supporting the development:
| Platform | Support Link |
|---|---|
| Global & India | Support via Razorpay |
Scan to support via UPI (India Only):
Connect with the architect: Abhishek Jaiswal on LinkedIn
- Downloads last month
- 327
Hardware compatibility
Log In to add your hardware
4-bit
5-bit
6-bit
8-bit
16-bit
Model tree for CelesteImperia/Phi-3.5-mini-instruct-Platinum-GGUF
Base model
microsoft/Phi-3.5-mini-instruct