MetaLlama 3.1 8B Instruct on NVIDIA GeForce RTX 4090
Yes — RTX 4090 runs Llama 3.1 8B Instruct excellently at Q8_0 — 95 tok/s. 24 GB VRAM with plenty of headroom.
Model Size
8.03B
Device VRAM
24 GB
Bandwidth
1008 GB/s
Quantization
Q8_0
Performance by Quantization
OwnRig currently has one published compatibility entry for Llama 3.1 8B Instruct on NVIDIA GeForce RTX 4090 at Q8_0. This is the best supported pairing we can stand behind today.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q8_0 | 95 tok/s | 65ms | ✓ Yes | Excellent | benchmarked |
Notes
Q8_0
The 4090 makes 8B models feel instant. 1008 GB/s bandwidth is overkill for this model size.
With nomic-embed running concurrently, speed drops ~5% to ~90 tok/s. Total VRAM: 9.3GB, leaving 14.7GB free.
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. General capabilities with capable 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 4090
NVIDIA GeForce RTX 4090 has 24 GB at 1008 GB/s. Street price: $1,799.
See all models NVIDIA GeForce RTX 4090 can run →Builds with NVIDIA GeForce RTX 4090
Source: llama.cpp benchmark thread (2026-01-15)
Performance varies by driver version, inference engine, quantization method, context length, and system configuration. Figures shown are estimates based on community benchmarks and may not reflect your exact setup. Product names are trademarks of their respective owners. OwnRig is independent and not affiliated with any hardware or AI model provider.