The datasheet's "Tensor Performance 634.0 TFLOPS" is footnoted "Effective FP8 teraFLOPS (TFLOPS) using sparsity", so the dense FP8 figure this catalog compares is half of it, 317.0. That is exactly 8x the card's FP32, the FP16-accumulate path, and it is the same convention already used on the RTX 6000 Ada and RTX 5000 Ada rows.

NVIDIA RTX 4500 Ada 24GB

Ada Lovelace PCIe 2023 4nm Spec confidence: Official
FP8 (dense)
317
TFLOPS
FP32
40
TFLOPS
VRAM
24 GB
GDDR6
Bandwidth
432 GB/s
memory
TDP
210 W
188.6 TFLOPS/kW

Overview

The NVIDIA RTX 4500 Ada Generation sits between the RTX 5000 Ada and the RTX 4000 Ada, with 7,680 CUDA cores, 240 fourth-generation Tensor cores and 60 third-generation RT cores, and 24 GB of GDDR6 with ECC on a 192-bit bus at 432 GB/s. At 210 W it delivers 39.6 TFLOPS of FP32 and 317 dense TFLOPS of FP8, the precision that made Ada worth having for inference. The 24 GB frame buffer is the practical draw: it is the smallest Ada professional card that will hold a 30B-parameter model at 8-bit weights, and it does it in a dual-slot board that most workstations can cool without changes. Ada tensor cores added FP8 over Ampere and kept BF16, TF32, FP16, INT8 and INT4, though NVIDIA publishes only the FP8 headline for this card.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
64-bit floating point
0.6TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
40TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP8
8-bit floating point
317TFLOPS 634TFLOPS
FP6
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP4
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
INT8
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision

Specifications

Architecture

Ada Lovelace

Form Factor

PCIe

Launch Year

2023

Process Node

4nm

Memory

24 GB GDDR6

Bandwidth

432 GB/s

TDP

210 W

Max Power

210 W

CUDA Cores

7,680

Spec Confidence

Official

Full Specifications

Compute Engine
CUDA Cores 7,680
Tensor Cores 240 (4th Gen)
Streaming Multiprocessors 60
Memory
VRAM 24 GB
Memory Type GDDR6
Bandwidth 432 GB/s
Interface Width 192-bit
Interconnect & I/O
PCIe 4.0 x16
Power & Thermal
TDP 210 W
Max Board Power 210 W
Cooling Active
Enterprise Features
ECC Memory Yes
Sparsity Yes
Compute APIs CUDA 12.2, OpenCL 3.0, DirectCompute
Physical & Media
Card Length 266.7 mm
Width 2-slot
NVENC Engines 2
NVDEC Engines 2
General
Form Factor PCIe
Architecture Ada Lovelace
Process Node 4nm
Launch Year 2023

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
2023-10-01

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
3 of 9
Normalization
All values in TFLOPS

Rent this GPU

$0.23 / GPU-hr

Lowest of 1 live listing across 1 provider

See all rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Similar GPUs

NVIDIA L40S 48GB

PCIe · 2023
FP32: 92 TFLOPS
Compare vs L40S

NVIDIA L40 48GB

PCIe · 2023
FP32: 91 TFLOPS
Compare vs L40

Frequently Asked Questions

How many TFLOPS does the NVIDIA RTX 4500 Ada have?

The NVIDIA RTX 4500 Ada delivers 40 TFLOPS FP32 and 317 TFLOPS FP8 at peak. Flopper does not currently have a verified FP16 throughput figure for it.

What is the power consumption of the NVIDIA RTX 4500 Ada?

The NVIDIA RTX 4500 Ada has a TDP (Thermal Design Power) rating of 210 watts.

How much memory does the NVIDIA RTX 4500 Ada have?

The NVIDIA RTX 4500 Ada is equipped with 24 GB of VRAM with 432 GB/s of memory bandwidth.

What architecture is the NVIDIA RTX 4500 Ada based on?

The NVIDIA RTX 4500 Ada is based on the Ada Lovelace architecture, launched in 2023.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs