Zum Hauptinhalt springen
Local Engine Ready

RTX 4080 Super

68 AI models fit in 16 GB VRAM at Q4 native. 15 more run with CPU offloading. Real benchmarks below.

68 Compatible Models
15 with offloading
16GB GDDR6X
320W TDP
Javier Morales
Javier Morales AI Hardware Specialist — 8 years experience
GitHub: github.com/javier-morales-ia

llama.cpp 0.2.x · CUDA 12 · Ubuntu 22.04 · Prices verified on Amazon · methodology →

RTX 4080 Super

Execution Context

BRAND NVIDIA
BANDWIDTH 736 GB/s
FP16 TFLOPS 52.2
AI SCORE 78 / 140
PRICE BAND Oberklasse
MSRP (LAUNCH) $999
AI SCORE (RELATIVE)
78 /140
Entry Value Mid High Flagship
MSRP at launch: $999 Current street price varies on Amazon

Aktuelles Angebot prüfen

Amazon affiliate link for RTX 4080 Super

Einige Links sind Amazon-Partnerlinks. Wir koennen ohne Mehrkosten fuer Sie eine Provision erhalten. Amazon-Cookies koennen nach Ihrem Klick bis zu 24 Stunden bestehen.

Prime

Full Specifications

NVIDIA · 2024-01

VRAM 16 GB GDDR6X
Bandwidth 736 GB/s
FP16 TFLOPS 52.2
AI Score 78 / 140
CUDA Cores 10,240
Tensor Cores 320
TDP 320 W
PCIe Gen 4
Slots 3
Power Connector 16-pin
Price Band Oberklasse
Released 2024-01

AI Benchmarks

Real inference measurements — llama.cpp Q4_K_M

Task Result
Llama 1B Q4 386 tok/s
Llama 3B Q4 155 tok/s
Llama 7B Q4 72 tok/s
Llama 13B Q4 34 tok/s
Llama 30B Q4 VRAM N/A
Llama 70B Q4 Offload or multi-GPU
Stable Diffusion 512px 3s / img
Whisper Large RTF 0.4x

RTF < 1.0 = faster than real time. For Stable Diffusion and Whisper lower is better; for tokens/s higher is better.

Compare RTX 4080 Super with another GPU

Is an upgrade worth it? Compare specs and real benchmarks side by side.

Open comparator →
Quick comparisons: vs M3 Ultra · vs RTX 5070 Ti

Compatible AI Models — RTX 4080 Super

68 models run fully in VRAM · 15 with CPU offloading

Show all 68 compatible models →

Also runs with CPU offloading (15)

RTX 4080 Super · Amazon

GPU-Preise schwanken haeufig zwischen Haendlern. Pruefen Sie vor dem Kauf das aktuelle Angebot.

Aktuelles Angebot prüfen

Einige Links sind Amazon-Partnerlinks. Wir koennen ohne Mehrkosten fuer Sie eine Provision erhalten. Amazon-Cookies koennen nach Ihrem Klick bis zu 24 Stunden bestehen.

RTX 4080 Super for Local AI

La RTX 4080 Super con 16GB de GDDR6X es una opción equilibrada para IA local. Puede correr modelos 7B y 13B en Q4 sin problemas, y modelos 7B en FP16. Es ideal para usuarios que quieren una experiencia fluida con los modelos más populares como Llama 3.1 8B, Mistral 7B, Phi-4 o Qwen2.5 14B.

Los benchmarks reales muestran 72 tokens/segundo en Llama 7B Q4. Para generación de imágenes, Stable Diffusion XL y SD 3.5 Medium caben perfectamente. Whisper Large V3 para transcripción de audio también es compatible.

Con 16GB de VRAM tienes acceso a 68 modelos completos y 15 adicionales con offloading parcial. Usa la calculadora de VRAM para ver las opciones de cuantización disponibles para cada modelo.

Plan your full AI build

RTX 4080 Super · 16 GB VRAM — configure PSU, RAM, storage and check compatible models.

Configure your build →

Not sure which model to run on your RTX 4080 Super?

The VRAM calculator tells you exactly which quantization you need.

Open calculator →

Get the best price for RTX 4080 Super

Open Amazon with our affiliate link and check availability, variants, and current deals.

Get Best Price →