Leaderboard Β· All comparisons Β· LynxBenchAI 26Q3 Index

NVIDIA GeForce RTX 4080 vs NVIDIA GeForce RTX 5080 for AI

Device pages: NVIDIA GeForce RTX 4080 Β· NVIDIA GeForce RTX 5080

On the LynxBenchAI 26Q3 Index, the NVIDIA GeForce RTX 4080 scores 1.08x the NVIDIA GeForce RTX 5080 overall (Training 0.89x, Inference 0.96x, Compute 2.17x). It leads on Compute and trails on Training and Inference.

Scores on the LynxBenchAI 26Q3 Index

Ratio = the RTX 4080's score divided by the RTX 5080's
ScoreRatioRTX 4080RTX 5080
Overall (GT)1.08x338.13311.78
Training0.89x250.74282.72
Inference0.96x487.16509.65
Compute2.17x290.01133.53

When to Choose Which

  • Choose the RTX 5080 for training (1.13x the RTX 4080).
  • Level on inference: the two are within 10% (0.96x).
  • Choose the RTX 4080 for raw compute (2.17x the RTX 5080).

From the LynxBenchAI 26Q3 Index category scores, which are not raw throughput; a lead under 10% counts as level.

Per Model and Precision

LynxBenchAI score per model and precision; ratio = RTX 4080 / RTX 5080
ModelPrecisionRatioRTX 4080RTX 5080
compute
Dense MatMulBF161.04x193.72186.38
Dense MatMulFP161.02x194.44190.03
Dense MatMulFP320.99x193.11194.65
Dense MatMulFP640.83x4.855.81
Sparse MatMulFP324.36x618.18141.84
Sparse MatMulFP644.52x475.98105.29
inference
DistilBERTFP160.93x679.07733.41
DistilBERTFP81.05x507.14482.05
DistilBERTINT81.01x459.23455.48
DistilGPT2FP161.11x633.84572.39
DistilGPT2FP81.22x412.17336.92
DistilGPT2INT81.19x410.40343.60
Flan-T5 SmallFP160.89x909.891018.11
Flan-T5 SmallFP80.97x656.48673.82
Flan-T5 SmallINT80.87x483.70558.02
GraphGPS (Peptides)FP160.81x237.93293.75
GraphGPS (Peptides)FP80.81x120.37148.59
GraphGPS (Peptides)INT80.83x120.89144.79
MobileNetV3FP160.84x1348.901598.29
MobileNetV3INT81.39x353.93254.00
training
DistilBERTBF160.82x305.83372.87
DistilBERTFP320.90x400.90443.76
DistilGPT2BF160.91x180.52198.02
DistilGPT2FP320.87x301.44348.25
Flan-T5 SmallBF160.80x205.83257.49
Flan-T5 SmallFP320.82x346.81424.46
GraphGPS (Peptides)BF161.00x162.59163.11
GraphGPS (Peptides)FP320.85x129.29152.72
MobileNetV3BF160.99x226.33229.48
MobileNetV3FP320.95x338.31356.54

Related Comparisons

Each figure is the first device's overall score divided by the second's.

Measure your own machine: pip install lynxbench-ai Β· Install guide β†’

How to Read These Results

A LynxBenchAI score is sustained throughput on a fixed catalogue of PyTorch models, measured after a warm-up at a batch size raised until throughput stops improving, so it is what the machine holds under load, not its best instant. Training, Inference and Compute are three separate scores; GT combines them (Training and Inference weigh 0.4 each, Compute 0.2).

Every run uses the stock, publicly released PyTorch stack, so a result is broadly indicative of the out-of-the-box PyTorch experience on that hardware, and so of custom PyTorch models too. It does not run your own model, and it is not a vendor-tuned peak like an MLPerf submission.

Scores compare only within one Index: the 26Q3 Index fixes the models, precisions, correctness thresholds and scoring, and a later Index may change them. Each test is one timed window, the scores have no 0-100 scale, and a precision a device cannot run scores zero rather than being skipped.

LynxBenchAI is built by TechnoLynx, which the European Commission engaged as an independent expert advisor on AI performance and benchmarking for AI Gigafactories under the EuroHPC Joint Undertaking (case study).

Questions

Is the NVIDIA GeForce RTX 4080 faster than the NVIDIA GeForce RTX 5080 for AI?
On the LynxBenchAI 26Q3 Index, the NVIDIA GeForce RTX 4080 scores 1.08x the NVIDIA GeForce RTX 5080 overall (Training 0.89x, Inference 0.96x, Compute 2.17x). It leads on Compute and trails on Training and Inference.
Should I choose the NVIDIA GeForce RTX 4080 or the NVIDIA GeForce RTX 5080?
Choose the RTX 5080 for training (1.13x the RTX 4080). Level on inference: the two are within 10% (0.96x). Choose the RTX 4080 for raw compute (2.17x the RTX 5080). These follow the LynxBenchAI 26Q3 Index category scores, which are not raw throughput.
Where is the difference between the NVIDIA GeForce RTX 4080 and the NVIDIA GeForce RTX 5080 largest?
On the LynxBenchAI 26Q3 Index, the ratio is highest on Sparse MatMul at FP64 (Compute), where the NVIDIA GeForce RTX 4080 scores 4.52x the NVIDIA GeForce RTX 5080, and lowest on Flan-T5 Small at BF16 (Training), at 0.80x.
What does this result tell me about my own PyTorch model?
Every run uses the stock, publicly released PyTorch stack, so a result is broadly indicative of the out-of-the-box PyTorch experience on that hardware, and so of custom PyTorch models too. It does not run your own model, and it is not a vendor-tuned peak like an MLPerf submission.

The Personal Edition submits your benchmark results to TechnoLynx's servers as part of participating in the public leaderboard. If that isn't acceptable for your use case, contact TechnoLynx about Press, Pro, or Enterprise licenses.

Get Started