THIS IS A TURING CARD WITH NO TENSOR CORES AND NO RT CORES, AND THAT IS NOT AN OMISSION. NVIDIA's own compare table prints "Ray Tracing Cores -" and "Tensor Cores -" for the whole 16 series, and the product page prints "Real-Time Ray Tracing No" and "Giga Rays/s NA". So tensor_core_count and tensor_core_gen are deliberately NULL, and the FP16 figure below is the SHADER rate with is_tensor_core = 0, not a tensor-core rate. That is also why there is no TF32, BF16, FP8, INT8 or INT4 row here where the RTX 20-series rows in this catalog have them. FP32 5.44 IS DERIVED from the published 1,536 CUDA cores and 1,770 MHz boost clock; NVIDIA publishes no TFLOPS figure for any 16-series card, hence spec_confidence vendor_claimed. FP16 at 2x and FP64 at 1/32 are the Turing ratios this catalog already applies to its other Turing rows. REFUSED: transistor count and die size, which NVIDIA never published for TU116 (its Turing whitepaper covers TU102, TU104 and TU106 only); l2_cache_mb, likewise; INT8 and INT4, since the integer tensor path this generation is known for is exactly what TU116 lacks.

NVIDIA GeForce GTX 1660 Ti 6GB

Turing PCIe 2019 12nm Spec confidence: Vendor claimed
FP32
5.4
TFLOPS
Memory
6 GB
GDDR6
Bandwidth
288 GB/s
memory
TDP
120 W
45.3 FP32 dense TFLOPS/kW of TDP
Download datasheet

Overview

The GeForce GTX 1660 Ti is Turing with the Turing removed: the same 12 nm generation and the same SM design as the RTX 20 series, but on a TU116 die that ships with no Tensor Cores and no RT Cores at all. What survives is the part that matters for cheap compute, a full-rate half-precision shader path, so its 1,536 CUDA cores at a 1,770 MHz boost give 5.44 TFLOPS FP32 and 10.87 TFLOPS FP16, twice the single-precision rate rather than the fraction every Pascal card manages. Against an RTX 2060 of the same era that is a real loss, since the 2060's Tensor Cores take FP16 to four times the shader rate and add an INT8 path entirely, but against a GTX 1070 it is the first NVIDIA consumer card below the RTX line where running a model in half precision is actually faster. It pairs that with 6 GB of GDDR6 at 12 Gbps for 288 GB/s inside 120 W, which is the frame buffer that decides most of what it can do.

Performance

Peak theoretical throughput by precision type

PrecisionPeak
FP64
64-bit floating point
0.2TFLOPS
FP32
32-bit floating point
5.4TFLOPS
TF32
No verified data available
BF16
No verified data available
FP16
16-bit floating point
11TFLOPS
FP8
No verified data available
FP6
No verified data available
FP4
No verified data available
INT8
No verified data available

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.

The GeForce GTX 1660 Ti 6GB in the GPU landscape

Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper

06001,2001,8002,4003,0000 W250 W500 W750 W1000 W1250 W1500 WInstinct MI355XGeForce GTX 1660 Ti 6GB

Higher and further left is better: more half-precision throughput for less power.

Specifications

Architecture

Turing

Form Factor

PCIe

Launch Year

2019

Process Node

12nm

Memory

6 GB GDDR6

Bandwidth

288 GB/s

TDP

120 W

Max power

120 W

CUDA Cores

1,536

Spec Confidence

Vendor claimed

Full Specifications

Compute Engine
CUDA Cores 1,536
Streaming Multiprocessors 24
Base Clock 1.50 GHz
Boost Clock 1.77 GHz
Memory
Memory 6 GB
Memory Type GDDR6
Bandwidth 288 GB/s
Interface Width 192-bit
Memory Clock 12 GT/s
Interconnect & I/O
PCIe 3.0 x16
Power & Thermal
TDP 120 W
Max power 120 W
Power Connector 8-pin
Cooling Active
Enterprise Features
ECC Memory No
Sparsity No
Compute APIs CUDA, DirectX 12, Vulkan, OpenGL
Physical & Media
Card Length 145 mm
Width 2-slot
General
Form Factor PCIe
Architecture Turing
Process Node 12nm
Launch Year 2019

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
NVIDIA
Published
No verified data available

Data Quality

Spec confidence
Vendor claimed
Clock basis
Boost
Core precisions with figures
3 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 Providers

Similar GPUs

Frequently Asked Questions

How many TFLOPS does the NVIDIA GeForce GTX 1660 Ti have?

The NVIDIA GeForce GTX 1660 Ti delivers 5.4 TFLOPS FP32 and 11 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 GTX 1660 Ti?

The NVIDIA GeForce GTX 1660 Ti has a TDP (Thermal Design Power) rating of 120 watts.

How much memory does the NVIDIA GeForce GTX 1660 Ti have?

The NVIDIA GeForce GTX 1660 Ti is equipped with 6 GB of memory with 288 GB/s of memory bandwidth.

What architecture is the NVIDIA GeForce GTX 1660 Ti based on?

The NVIDIA GeForce GTX 1660 Ti is based on the Turing architecture, launched in 2019.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs