RTX 4070

NVIDIA · 12 GB · 504 GB/s

3 models3 model configurations3 exact recipes

Representative image: 4070-family og:image depicts the 4070 SUPER FE; representative for the 4070 (shares the same FE chassis). — 4070-family og:image depicts the 4070 SUPER FE; representative for the 4070 (shares the same FE chassis). NVIDIA ›

Model configurations

Each card is a model and hardware setup, including any documented host RAM, storage or CPU requirements. Select a recipe to update its speeds, context and exact command.

Search configurations

Sort by

3 configurations · 3 recipes

Gemma 4 12B

RTX 407012 GB1 card per machine · 1 machine

TabbyAPIEXL3 4 bpw

54.8decode tok/s

1,211prefill tok/s

64Kcontext

Tested on this card · Sep 26, 2026

Full recipe ›

Qwen3.8-27B

RTX 407012 GB1 card per machine · 1 machine

TabbyAPIEXL3 2.2 bpw

54.4decode tok/s

681prefill tok/s

32Kcontext

Tested on this card · Sep 26, 2026

Full recipe ›

Qwen3.8-Flash-Next

RTX 407012 GB1 card per machine · 1 machine

Host RAM: ≥ 55 GB free
Disk: ≥ 86 GB NVMe

SGLangEXL3 3.05 bpw

35.4decode tok/s

1,460prefill tok/s

128Kcontext

Tested on this card · Oct 16, 2026

Full recipe ›

Speeds are the selected recipe’s recorded measurements, not estimates; “—” means unmeasured. Evidence distinguishes checks on this card, sibling-card checks, earlier checks and publisher reports.

Hardware

Memory	12 GB GDDR6X
Bandwidth	504 GB/s
Architecture	NVIDIA Ada Lovelace
CUDA cores	5,888

Precision	Dense TFLOPS	2:4 sparse
FP32	29	–

MSRP	$599
Street price	$1,087
Availability	In stock

Listing	Price	Seen	Where
NVIDIA GeForce RTX 4070 Founders Edition Graphics Card 12GB GDDR6X	$1,087	2026-08-31	newegg ›
MSRP	$599	2023-04-13	NVIDIA ›

Sources: NVIDIA, NVIDIA (2), NVIDIA (3), newegg