OwnRig

Llama 3.1 8B Instruct on NVIDIA GeForce RTX 5080

NVIDIA GeForce RTX 5080 runs Llama 3.1 8B Instruct excellently at 92 tok/s at Q8_0. This is a strong pairing.

Model Size

8.03B

Device VRAM

16 GB

Bandwidth

960 GB/s

Quantizations Tested

1

Performance by Quantization

Each row shows Llama 3.1 8B Instruct performance at a different quality level on NVIDIA GeForce RTX 5080.

QuantizationSpeedTTFTFits in VRAMRatingConfidence
Q8_092 tok/s65ms✓ YesExcellentestimated

Notes

Q8_0

RTX 5080 (960 GB/s) similar to 4090. Full Q8 quality.

About Llama 3.1 8B Instruct

Llama 3.1 8B Instruct (8.03B) is a chat, coding, ai coding, reasoning, multi-purpose model. Best-in-class 8B model. Strong general capabilities with excellent coding support. The go-to small model for local inference — fast, accurate, and well-supported across all inference engines.

View all Llama 3.1 8B Instruct hardware options →

About NVIDIA GeForce RTX 5080

NVIDIA GeForce RTX 5080 has 16 GB at 960 GB/s. Street price: $1,099.

See all models NVIDIA GeForce RTX 5080 can run →

Source: Community benchmarks and estimated performance (2026-03-01)

Data last updated: 2026-03-01