Can I run Llama-3.1-70B-Instruct on a NVIDIA RTX 6000 Ada?
Yes.Runs
You can run this with Q3 quantization. On a NVIDIA RTX 6000 Ada (48 GB VRAM), Llama-3.1-70B-Instruct needs about 37 GB (Q3) and should generate at roughly ~22 tok/s. Recommended: Run on your GPU with Q3.
meta-llama/Llama-3.1-70B-InstructWhy
- Full precision needs ~170 GB which exceeds your accelerator, but 3-bit (Q3_K) needs only ~37 GB and fits.
How to run Llama-3.1-70B-Instruct on a RTX 6000 Ada
Run with Ollama (easiest)
ollama run llama3.1:70bOr with llama.cpp
llama-cli -hf meta-llama/Llama-3.1-70B-Instruct -p "Hello"Serve with vLLM (OpenAI-compatible API)
# Fast production serving on http://localhost:8000/v1
vllm serve meta-llama/Llama-3.1-70B-InstructPython (Transformers)
from transformers import pipeline, BitsAndBytesConfig
# pip install transformers accelerate bitsandbytes
quant = BitsAndBytesConfig(load_in_4bit=True)
pipe = pipeline("text-generation", model="meta-llama/Llama-3.1-70B-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 Llama-3.1-70B-Instruct well is the NVIDIA H100 (80GB) (80 GB).