NVIDIA · Turing (derived from codename prefix) · 2019
GeForce RTX 2070 — TU106
GeForce RTX 2070 — TU106: 8 GB GDDR6, 115 W, 2,304 CUDA cores. Launched January 29, 2019.
- Memory
- 8 GB
- Power
- 115 W
- Bus width
- 256 bit
Memory8 GB
33 of 77 in GeForce RTX
Power115 W
55 of 66 by higher draw in GeForce RTX
CUDA cores2,304
67 of 78 in GeForce RTX
Current market
Affiliate links — we may earn a commission.
Specifications
Memory
- Capacity
- 8 GB
- Memory type
- GDDR6
- Bus width
- 256 bit
- Bandwidth
- 448 GB/s
Compute
- CUDA cores
- 2,304
- SM count
- 36
- TMUs
- 144
- ROPs
- 64
- FP32
- 6,636 GFLOPS
- FP64
- 207.4 GFLOPS
- Pixel fill rate
- 92.16 GP/s
- Texture fill rate
- 207.4 GT/s
Clocks
- Base clock
- 1,215 MHz
- Boost clock
- 1,440 MHz
Board
- TDP
- 115 W
- Bus interface
- PCIe 3.0 x16
- Form factor
- mobile
Physical
- Architecture
- Turing (derived from codename prefix)
- Codename
- TU106
- Process
- 12 nm (mapped from a reviewed process name)
- Transistors
- 10.8 billion
- Die size
- 445 mm²
- Series
- GeForce RTX
- Launch date
- January 29, 2019
Not recorded by the source: generation, memory clock, external power required and launch msrp.
Die figure
Silicon, made visible
Die area is drawn against the largest recorded die in the catalog. Shader dots use a labeled, card-specific ratio so both early and modern GPUs remain legible.
Catalog context
Within its series, the GeForce RTX 2070 ranks 54th of 66 for FP32 throughput and 33rd of 77 for memory capacity. Its 115 W TDP is above the 75 W PCIe slot-power ceiling.
Source: Wikipedia, revision 1369488585 · catalog loaded 2026-08-16