Llama 3.3 70B Instruct (Q5_K_M) — 47.4 GBon NVIDIA L40S

Q5_K_M NVIDIA L40S

Overview

Llama 3.3 70B Instruct is a 70B parameter dense language model by Meta, with code, multilingual, tool-calls capabilities. It supports a context window of up to 131,072 tokens.

Llama 3.3 70B Instruct is a 70-billion-parameter dense transformer model from Meta, optimized for instruction following, code generation, and multilingual conversation. It delivers performance competitive with larger models in the Llama family while remaining practical for single-node GPU deployments. The model supports tool calling and eight languages including English, French, Spanish, and German. With a 128K context window and grouped-query attention, it quantizes efficiently down to Q4 levels for self-hosted inference on consumer hardware.

At Q5_K_M quantization (medium quality tier), the model weighs 46.52 GB. This fits within the 48 GB of VRAM on NVIDIA L40S, enabling full GPU inference.

The NVIDIA L40S is a datacenter GPU with 48 GB of GDDR6 VRAM and 864 GB/s memory bandwidth. It delivers 362 FP16 TFLOPS with Ada Lovelace architecture. A versatile GPU for AI inference, training, and graphics workloads. Handles quantized models up to 40B parameters comfortably.

Hardware Requirements

Model size	46.52 GB
VRAM available	48 GB
VRAM used	47.4 GB
Min RAM required	0.6 GB
GPU layers	79 / 80
Context size	710
Backend	cuda13
Flash attention	Yes

Performance Notes

Reduced performance

1 of 80 layers run on CPU. Your system needs at least 0.6 GB RAM. If system RAM is insufficient, data will be read from disk on demand, with significant performance impact.

Shorter conversations

Context size has been reduced due to memory constraints. This may affect the model's ability to handle long conversations.

Deploy

Server Local

Prerequisites

Ensure your GPU nodes are prepared with the NVIDIA container toolkit:

ansible-playbook prositronic.infra.nvidia_container_toolkit

Command

helmfile --state-values-file <(curl -s https://www.prositronic.eu/values/llama-3-3-70b-instruct/q5_k_m/nvidia-l40s.yaml) apply

Generated values.yaml

/values/llama-3-3-70b-instruct/q5_k_m/nvidia-l40s.yaml

Loading values…

Install llama.cpp

Install llama.cpp from the official build instructions for your platform and backend.

Download Model

curl -L -o llama-3-3-70b-instruct.gguf "https://huggingface.co/bartowski/Llama-3.3-70B-Instruct-GGUF/resolve/main/Llama-3.3-70B-Instruct-Q5_K_M/Llama-3.3-70B-Instruct-Q5_K_M-00001-of-00002.gguf"

Start Server

llama-server \
  -m llama-3-3-70b-instruct.gguf \
  --n-gpu-layers 79 \
  --ctx-size 710 \
  --flash-attn

Verify

curl http://localhost:8080/health

Explore

• Configure a deployment
• Browse models
• Browse hardware

Other quantizations

• Llama 3.3 70B Instruct (FP16)
• Llama 3.3 70B Instruct (Q8_0)
• Llama 3.3 70B Instruct (Q4_K_M)

Other hardware

• Llama 3.3 70B Instruct on CPU Only
• Llama 3.3 70B Instruct on NVIDIA A100 80GB
• Llama 3.3 70B Instruct on NVIDIA H100

Frequently Asked Questions

How much VRAM does Llama 3.3 70B Instruct (Q5_K_M) need?

The Q5_K_M quantization of Llama 3.3 70B Instruct requires 46.52 GB. 79 of 80 layers fit in the 48 GB of VRAM on NVIDIA L40S; remaining layers are offloaded to CPU.

Can I run Llama 3.3 70B Instruct on NVIDIA L40S?

Yes, with reduced performance. NVIDIA L40S can run Llama 3.3 70B Instruct (Q5_K_M), but only 79 of 80 layers fit in VRAM. The rest are offloaded to CPU.

What is quantization?

Quantization reduces a model's numerical precision from its original floating-point format to a more compact representation. This shrinks the file size and VRAM footprint, making it possible to run large models on consumer hardware. The trade-off is a small reduction in output quality. Q5_K_M compresses Llama 3.3 70B Instruct from its original size down to 46.52 GB.

What quantization should I choose for Llama 3.3 70B Instruct?

Q5_K_M is a medium-quality quantization. Higher-quality quants (Q8, Q6) preserve more model accuracy but need more VRAM. Lower quants (Q4, Q3, Q2) reduce VRAM usage at the cost of some quality. Choose based on your available hardware and quality requirements.

What is flash attention and why is it enabled?

Flash attention is a memory-efficient algorithm that speeds up the attention mechanism in transformer models. It reduces VRAM usage during inference by avoiding the materialisation of the full attention matrix. For Llama 3.3 70B Instruct on NVIDIA L40S, flash attention is enabled to maximise context length and throughput within the available 48 GB of VRAM.

Why are some layers offloaded to CPU?

NVIDIA L40S has 48 GB of VRAM, but Llama 3.3 70B Instruct (Q5_K_M) requires approximately 46.52 GB. Only 79 of 80 layers fit in VRAM; the remaining layers run on CPU, which is slower but still functional.

How do I run Llama 3.3 70B Instruct (Q5_K_M) with Ollama?

Run ollama run llama3.3:70b-instruct-q5_k_m to start Llama 3.3 70B Instruct (Q5_K_M). Ollama handles downloading the model weights automatically on first run.

Last updated: March 5, 2026