Llama 3.1 70B Instruct on NVIDIA GeForce RTX 5090
NVIDIA GeForce RTX 5090 handles Llama 3.1 70B Instruct well at 14 tok/s at Q4_K_M. A solid choice for this model.
Model Size
70.6B
Device VRAM
32 GB
Bandwidth
1792 GB/s
Quantizations Tested
1
Performance by Quantization
Each row shows Llama 3.1 70B Instruct performance at a different quality level on NVIDIA GeForce RTX 5090.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q4_K_M | 14 tok/s | 1100ms | ✓ Yes | Good | estimated |
Notes
Q4_K_M
32GB VRAM fits 70B Q4. 5090 bandwidth makes it usable without offloading.
About Llama 3.1 70B Instruct
Llama 3.1 70B Instruct (70.6B) is a chat, coding, ai coding, reasoning, multi-purpose model. Frontier-class open model. Approaches GPT-4 quality on many benchmarks. Requires significant VRAM — 48GB+ recommended for usable quantizations. Excellent for serious local deployment.
View all Llama 3.1 70B Instruct hardware options →About NVIDIA GeForce RTX 5090
NVIDIA GeForce RTX 5090 has 32 GB at 1792 GB/s. Street price: $2,199.
See all models NVIDIA GeForce RTX 5090 can run →Builds with NVIDIA GeForce RTX 5090
Source: Community benchmarks and estimated performance (2026-03-01)
Data last updated: 2026-03-01