Can I run Phi-3-mini-4k-instruct on a NVIDIA RTX 2080 Ti?
Yes.Runs
You can run this with Q8 quantization. On a NVIDIA RTX 2080 Ti (11 GB VRAM), Phi-3-mini-4k-instruct needs about 5.6 GB (Q8) and should generate at roughly ~92 tok/s. Recommended: Run on your GPU with Q8.
microsoft/Phi-3-mini-4k-instructWhy
- Full precision needs ~10 GB which exceeds your accelerator, but 8-bit (INT8 / Q8) needs only ~5.6 GB and fits.
How to run Phi-3-mini-4k-instruct on a RTX 2080 Ti
Run with Ollama (easiest)
ollama run phi3Or with llama.cpp
llama-cli -hf microsoft/Phi-3-mini-4k-instruct -p "Hello"Serve with vLLM (OpenAI-compatible API)
# Fast production serving on http://localhost:8000/v1
vllm serve microsoft/Phi-3-mini-4k-instructPython (Transformers)
from transformers import pipeline, BitsAndBytesConfig
# pip install transformers accelerate bitsandbytes
quant = BitsAndBytesConfig(load_in_8bit=True)
pipe = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct", device_map="auto",
model_kwargs={"quantization_config": quant})
print(pipe("Hello", max_new_tokens=50))Want it to run comfortably at higher precision? The smallest GPU that runs Phi-3-mini-4k-instruct well is the NVIDIA RTX 5070 (12 GB).