NVIDIA GeForce RTX 2070 8GB
Overview
The GeForce RTX 2070 is the full TU106 die and the entry point of the first ray tracing generation. NVIDIA gave it an entire whitepaper appendix, so its precision ladder is transcribed rather than derived: 31.5 TFLOPS of FP16 tensor accumulating in FP32, 63 accumulating in FP16, and 126 TOPS of INT8, on 8 GB of GDDR6 at 448 GB/s. Unlike the larger Turing dies it has no NVLink at all, so there is no two-card option, and unlike the RTX 2070 SUPER that replaced it, it is a genuine TU106 rather than a cut TU104. Turing has no BF16, no TF32 and no structured sparsity, so everything here is dense.
Performance
Peak theoretical throughput by precision type
| Precision | Peak |
|---|---|
FP64 64-bit floating point | 0.2TFLOPS |
FP32 32-bit floating point | 7.9TFLOPS |
TF32 | No verified data available |
BF16 | No verified data available |
FP16 16-bit floating point | 32TFLOPS |
FP8 | No verified data available |
FP6 | No verified data available |
FP4 | No verified data available |
INT8 8-bit integer | 126TOPS |
INT4 4-bit integer | 252TOPS |
Every figure here is dense. The vendor documents no structured (2:4) sparsity mode for this part, so there is no second number to quote.
Per watt divides the dense figure by the 185 W board power. Where only a sparse figure exists, the cell says so.
The GeForce RTX 2070 8GB in the GPU landscape
Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper
Higher and further left is better: more half-precision throughput for less power.
Specifications
Architecture
Turing
Form Factor
PCIe
Launch Year
2018
Process Node
12nm
Memory
8 GB GDDR6
Bandwidth
448 GB/s
TDP
185 W
Max Power
185 W
Transistors
10.8 billion
CUDA Cores
2,304
Spec Confidence
Official
Full Specifications
| Compute Engine | |
|---|---|
| CUDA Cores | 2,304 |
| Tensor Cores | 288 (Turing) |
| Streaming Multiprocessors | 36 |
| Base Clock | 1.41 GHz |
| Boost Clock | 1.71 GHz |
| Chip Design | |
| Transistors | 10.8 billion |
| Die Size | 445 mm² |
| Process Node | 12nm |
| Memory | |
| VRAM | 8 GB |
| Memory Type | GDDR6 |
| Bandwidth | 448 GB/s |
| Interface Width | 256-bit |
| Memory Clock | 14 GT/s |
| Interconnect & I/O | |
| PCIe | 3.0 x16 |
| Cache | |
| L2 Cache | 4 MB |
| Power & Thermal | |
| TDP | 185 W |
| Max Board Power | 185 W |
| Cooling | Active |
| Enterprise Features | |
| ECC Memory | No |
| Sparsity | No |
| Compute APIs | CUDA, DirectX 12, Vulkan, OpenGL |
| Physical & Media | |
| Card Length | 228.6 mm |
| Width | 2-slot |
| General | |
| Form Factor | PCIe |
| Architecture | Turing |
| Process Node | 12nm |
| Launch Year | 2018 |
Datasheet & Resources
Flopper Datasheet
NVIDIA GeForce RTX 2070 8GB specifications, generated from our database. Printable.
Vendor product page
NVIDIA documentation for the GeForce RTX 2070 8GB
GeForce RTX 2070 product page (nvidia.com, archived 2018-11-09)
NVIDIA · 2018-08-20
NVIDIA Developer Documentation
Technical specs, programming guides
Data Provenance
Every figure traced to a source
Primary Source
- Publisher
- NVIDIA
- Published
- 2018-08-20
Data Quality
- Spec confidence
- Official
- Clock basis
- Boost
- Core precisions with figures
- 4 of 9
- Normalization
- All values in TFLOPS
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Frequently Asked Questions
How many TFLOPS does the NVIDIA GeForce RTX 2070 have?
The NVIDIA GeForce RTX 2070 delivers 7.9 TFLOPS FP32 and 32 TFLOPS FP16 at peak. Flopper does not currently have a verified FP8 throughput figure for it.
What is the power consumption of the NVIDIA GeForce RTX 2070?
The NVIDIA GeForce RTX 2070 has a TDP (Thermal Design Power) rating of 185 watts.
How much memory does the NVIDIA GeForce RTX 2070 have?
The NVIDIA GeForce RTX 2070 is equipped with 8 GB of VRAM with 448 GB/s of memory bandwidth.
What architecture is the NVIDIA GeForce RTX 2070 based on?
The NVIDIA GeForce RTX 2070 is based on the Turing architecture, launched in 2018.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.