OwnRig

DeepSeek V3 on NVIDIA GeForce RTX 4070 Ti 12GB

NVIDIA GeForce RTX 4070 Ti 12GB cannot run DeepSeek V3 at any quantization level. The 12 GB of VRAM is insufficient.

Model Size

671B

Device VRAM

12 GB

Bandwidth

504 GB/s

Quantizations Tested

1

Performance by Quantization

Each row shows DeepSeek V3 performance at a different quality level on NVIDIA GeForce RTX 4070 Ti 12GB.

QuantizationSpeedTTFTFits in VRAMRatingConfidence
Q2_K✗ OffloadNot Viableestimated

Notes

Q2_K

671B parameter model. Cloud/inference API only.

About DeepSeek V3

DeepSeek V3 (671B) is a chat, coding, ai coding, reasoning, multi-purpose model. Massive MoE model rivaling GPT-4 class. Only ~37B parameters active per token despite 671B total. Requires multi-GPU or very large unified memory (128GB+ Apple Silicon at Q2/Q3). Not for casual home use — included for completeness and to show what the high end looks like.

View all DeepSeek V3 hardware options →

About NVIDIA GeForce RTX 4070 Ti 12GB

NVIDIA GeForce RTX 4070 Ti 12GB has 12 GB at 504 GB/s. Street price: $749.

See all models NVIDIA GeForce RTX 4070 Ti 12GB can run →

Source: 671B MoE model requires hundreds of GB VRAM (2026-03-15)

Data last updated: 2026-03-15