GeForce GTX 1070 Ti vs GeForce GTX 1660 Ti: specs and benchmark comparison
Rendering and compute
depends on the taskBenchmark results disagree, so the choice depends on the specific task.
No comparable benchmarks yet: gaming: rasterization, gaming: ray tracing, power efficiency.
Advantages
Only measured and rated values: a noticeable benchmark gap and specs where more is better.
Where GeForce GTX 1070 Ti is better5
Blender Benchmark 2.93.1 8.4% lower78 s vs 84 s
Blender Benchmark 2.90 13.6% lower294 s vs 333 s
Blender Benchmark 2.90 14.8% lower503 s vs 578 s
Video memory 2 GB more8 GB vs 6 GB
Memory bus 64 bit more256 bit vs 192 bit
Where GeForce GTX 1660 Ti is better4
Blender Benchmark 4.5.0 27.2% faster753 samples/min vs 592 samples/min
Memory bandwidth 31.7 GB/s more288 GB/s vs 256.3 GB/s
Board power 60 W lower120 W (TDP) vs 180 W (TDP)
Launch price $120 lower$279 vs $399
Benchmarks
Results are compared within one benchmark, one version and the same conditions.
- A ahead by 8.4%
Blender Benchmark 2.93.1
GPU, bmw27 (CUDA) · s, lower is better
GeForce GTX 1070 Ti77.7 sGeForce GTX 1660 Ti84.2 sPosition among tested models (31, CUDA)
A · 78B · 840100200300Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce GTX 1070 Ti is ahead by 8.4% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
77.7 s · community measurement, sample 20 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 20 submissions (2 outliers excluded), interquartile range 76.96–78.05; submissions 2021-07-12 to 2022-02-25; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
84.2 s · community measurement, sample 43 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 43 submissions (0 outliers excluded), interquartile range 80.69–88.8; submissions 2021-07-14 to 2022-02-26; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- A ahead by 13.6%
Blender Benchmark 2.90
GPU, classroom (CUDA) · s, lower is better
GeForce GTX 1070 Ti293.5 sGeForce GTX 1660 Ti333.4 sPosition among tested models (25, CUDA)
A · 294B · 3330200400600800Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce GTX 1070 Ti is ahead by 13.6% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
293.5 s · community measurement, sample 75 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 75 submissions (3 outliers excluded), interquartile range 280.96–296.59; submissions 2020-09-02 to 2021-01-31; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
333.4 s · community measurement, sample 105 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 105 submissions (8 outliers excluded), interquartile range 325.46–343.87; submissions 2020-09-02 to 2021-02-04; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- 1.1% difference, within the threshold
Blender Benchmark 2.90
GPU, fishy_cat (CUDA) · s, lower is better
GeForce GTX 1070 Ti187.6 sGeForce GTX 1660 Ti185.6 sPosition among tested models (22, CUDA)
B · 186A · 1880200400600Each dot is one model’s median in this benchmark. The scale starts at zero, s. The 1.1% difference is below the noticeable threshold (5%).
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
187.6 s · community measurement, sample 56 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 56 submissions (3 outliers excluded), interquartile range 179.05–190.86; submissions 2020-09-02 to 2021-01-31; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
185.6 s · community measurement, sample 84 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 84 submissions (4 outliers excluded), interquartile range 182.06–195.79; submissions 2020-09-04 to 2021-02-04; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- 0.7% difference, within the threshold
Blender Benchmark 2.90
GPU, koro (CUDA) · s, lower is better
GeForce GTX 1070 Ti296 sGeForce GTX 1660 Ti298 sPosition among tested models (22, CUDA)
A · 296B · 2980200400600800Each dot is one model’s median in this benchmark. The scale starts at zero, s. The 0.7% difference is below the noticeable threshold (5%).
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
296 s · community measurement, sample 52 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 52 submissions (3 outliers excluded), interquartile range 283.33–300.59; submissions 2020-09-02 to 2021-01-31; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
298 s · community measurement, sample 77 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 77 submissions (6 outliers excluded), interquartile range 292.89–312.81; submissions 2020-09-04 to 2021-02-04; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- A ahead by 14.8%
Blender Benchmark 2.90
GPU, pavillon_barcelona (CUDA) · s, lower is better
GeForce GTX 1070 Ti502.9 sGeForce GTX 1660 Ti577.5 sPosition among tested models (23, CUDA)
A · 503B · 57805001,0001,500Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce GTX 1070 Ti is ahead by 14.8% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
502.9 s · community measurement, sample 52 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 52 submissions (3 outliers excluded), interquartile range 479.86–509.1; submissions 2020-09-02 to 2021-01-31; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
577.5 s · community measurement, sample 82 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 82 submissions (5 outliers excluded), interquartile range 564.82–600.01; submissions 2020-09-04 to 2021-02-04; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- 2.4% difference, within the threshold
Blender Benchmark 2.90
GPU, victor (CUDA) · s, lower is better
GeForce GTX 1070 Ti692.6 sGeForce GTX 1660 Ti709.3 sPosition among tested models (20, CUDA)
A · 693B · 70905001,0001,500Each dot is one model’s median in this benchmark. The scale starts at zero, s. The 2.4% difference is below the noticeable threshold (5%).
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
692.6 s · community measurement, sample 46 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 46 submissions (8 outliers excluded), interquartile range 663.89–711.48; submissions 2020-09-02 to 2021-01-31; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
709.3 s · community measurement, sample 77 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 77 submissions (7 outliers excluded), interquartile range 693.39–748.04; submissions 2020-09-04 to 2021-02-04; CUDA backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Render time of a single Cycles scene without synchronization. Median of submissions after filtering.
- B ahead by 27.2%
Blender Benchmark 4.5.0
GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) · samples/min, higher is better
GeForce GTX 1070 Ti591.6 samples/minGeForce GTX 1660 Ti752.7 samples/minPosition among tested models (67, OptiX)
A · 592B · 75305,00010,00015,000Each dot is one model’s median in this benchmark. The scale starts at zero, samples/min. NVIDIA GeForce GTX 1660 Ti is ahead by 27.2% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce GTX 1070 Ti
591.6 samples/min · community measurement, sample 24 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 24 submissions (1 outliers excluded), interquartile range 582.59–600.66; submissions 2025-07-19 to 2026-04-11; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce GTX 1660 Ti
752.7 samples/min · community measurement, sample 58 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 58 submissions (4 outliers excluded), interquartile range 713.01–780.43; submissions 2025-07-17 to 2026-07-24; OPTIX backend. Snapshot 2026-11-14.
difference = |A − B| ÷ the smaller value × 100%. The 5% noticeable threshold is a UX heuristic, not statistical significance.
Cycles render in Blender Benchmark. Median of the sum of samples per minute over three scenes after removing outliers.
Other comparable benchmarks (18)
| Benchmark | A | B | Difference |
|---|---|---|---|
| Blender Benchmark 2.81aGPU, bmw27 (CUDA) | 86.8 | 90.2 | A +4% |
| Blender Benchmark 2.82GPU, bmw27 (CUDA) | 88 | 91.7 | A +4.2% |
| Blender Benchmark 2.83GPU, bmw27 (CUDA) | 89.3 | 92.4 | A +3.4% |
| Blender Benchmark 2.90GPU, bmw27 (CUDA) | 96.2 | 98.6 | A +2.4% |
| Blender Benchmark 2.82GPU, classroom (CUDA) | 285.5 | 322.2 | A +12.9% |
| Blender Benchmark 2.83GPU, classroom (CUDA) | 286.1 | 328 | A +14.7% |
| Blender Benchmark 2.82GPU, fishy_cat (CUDA) | 182.2 | 173.3 | B +5.1% |
| Blender Benchmark 2.83GPU, fishy_cat (CUDA) | 180.9 | 177.1 | B +2.1% |
| Blender Benchmark 2.82GPU, koro (CUDA) | 338 | 284.9 | B +18.6% |
| Blender Benchmark 2.82GPU, pavillon_barcelona (CUDA) | 474.3 | 543.8 | A +14.7% |
| Blender Benchmark 2.83GPU, pavillon_barcelona (CUDA) | 485.9 | 546.7 | A +12.5% |
| Blender Benchmark 2.82GPU, victor (CUDA) | 714.3 | 704.2 | B +1.4% |
| Blender Benchmark 3.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 633.8 | 820.5 | B +29.5% |
| Blender Benchmark 3.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 635.3 | 825.9 | B +30% |
| Blender Benchmark 3.4.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 639.7 | 825.2 | B +29% |
| Blender Benchmark 3.5.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 662.4 | 862.4 | B +30.2% |
| Blender Benchmark 3.6.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 635.4 | 854.4 | B +34.5% |
| Blender Benchmark 4.0.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 596.6 | 781.6 | B +31% |
Each benchmark version has its own scale: numbers are compared only within a row.
Reference results (76)
| Benchmark | A | B | Difference |
|---|---|---|---|
| Blender Benchmark 2.81GPU, barbershop_interior (CUDA) | 802.3 | 1,104.2 | reference |
| Blender Benchmark 2.81aGPU, barbershop_interior (CUDA) | 802.9 | 1,122.6 | reference |
| Blender Benchmark 2.81GPU, bmw27 (CUDA) | 88.4 | 90.5 | reference |
| Blender Benchmark 2.91.2GPU, bmw27 (CUDA) | 89.7 | 94.2 | reference |
| Blender Benchmark 2.92GPU, bmw27 (CUDA) | 89.9 | 97.3 | reference |
| Blender Benchmark 2.90GPU, bmw27 (OPTIX) | 90 | 74.2 | reference |
| Blender Benchmark 2.91.2GPU, bmw27 (OPTIX) | 88.9 | 74.5 | reference |
| Blender Benchmark 2.92GPU, bmw27 (OPTIX) | 89.5 | 72.3 | reference |
| Blender Benchmark 2.93.1GPU, bmw27 (OPTIX) | 91.7 | 78 | reference |
| Blender Benchmark 2.81GPU, classroom (CUDA) | 286.5 | 331 | reference |
| Blender Benchmark 2.81aGPU, classroom (CUDA) | 284.4 | 333 | reference |
| Blender Benchmark 2.91.2GPU, classroom (CUDA) | 274.6 | 338.7 | reference |
| Blender Benchmark 2.92GPU, classroom (CUDA) | 280 | 341.5 | reference |
| Blender Benchmark 2.93.1GPU, classroom (CUDA) | 253.8 | 300.4 | reference |
| Blender Benchmark 2.90GPU, classroom (OPTIX) | 325.3 | 298 | reference |
| Blender Benchmark 2.92GPU, classroom (OPTIX) | 338.1 | 289.8 | reference |
| Blender Benchmark 2.93.1GPU, classroom (OPTIX) | 328.2 | 307.1 | reference |
| Blender Benchmark 2.81GPU, fishy_cat (CUDA) | 176.7 | 166.2 | reference |
| Blender Benchmark 2.81aGPU, fishy_cat (CUDA) | 178.2 | 168.3 | reference |
| Blender Benchmark 2.91.2GPU, fishy_cat (CUDA) | 176.6 | 169.5 | reference |
| Blender Benchmark 2.92GPU, fishy_cat (CUDA) | 178.1 | 183.9 | reference |
| Blender Benchmark 2.93.1GPU, fishy_cat (CUDA) | 181.7 | 169.5 | reference |
| Blender Benchmark 2.90GPU, fishy_cat (OPTIX) | 168.7 | 146.3 | reference |
| Blender Benchmark 2.92GPU, fishy_cat (OPTIX) | 175 | 158.6 | reference |
| Blender Benchmark 2.93.1GPU, fishy_cat (OPTIX) | 178.8 | 165.4 | reference |
| Blender Benchmark 2.81GPU, koro (CUDA) | 340.9 | 286.5 | reference |
| Blender Benchmark 2.81aGPU, koro (CUDA) | 341.7 | 288.5 | reference |
| Blender Benchmark 2.83GPU, koro (CUDA) | 327.8 | 302.4 | reference |
| Blender Benchmark 2.91.2GPU, koro (CUDA) | 291.9 | 275 | reference |
| Blender Benchmark 2.92GPU, koro (CUDA) | 295.3 | 286.2 | reference |
| Blender Benchmark 2.93.1GPU, koro (CUDA) | 300.5 | 269.1 | reference |
| Blender Benchmark 2.90GPU, koro (OPTIX) | 195.9 | 182.5 | reference |
| Blender Benchmark 2.92GPU, koro (OPTIX) | 281.2 | — | no pair |
| Blender Benchmark 2.93.1GPU, koro (OPTIX) | 206.8 | 230 | reference |
| Blender Benchmark 2.81GPU, pavillon_barcelona (CUDA) | 480.4 | 548.6 | reference |
| Blender Benchmark 2.81aGPU, pavillon_barcelona (CUDA) | 472.5 | 546.1 | reference |
| Blender Benchmark 2.91.2GPU, pavillon_barcelona (CUDA) | 471.7 | 528.2 | reference |
| Blender Benchmark 2.92GPU, pavillon_barcelona (CUDA) | 483.3 | 538.2 | reference |
| Blender Benchmark 2.93.1GPU, pavillon_barcelona (CUDA) | 447.2 | 481.4 | reference |
| Blender Benchmark 2.90GPU, pavillon_barcelona (OPTIX) | 475.9 | 429.1 | reference |
| Blender Benchmark 2.92GPU, pavillon_barcelona (OPTIX) | 491.7 | — | no pair |
| Blender Benchmark 2.93.1GPU, pavillon_barcelona (OPTIX) | 501.4 | 454.8 | reference |
| Blender Benchmark 2.81GPU, victor (CUDA) | 773 | 712.3 | reference |
| Blender Benchmark 2.81aGPU, victor (CUDA) | 750.1 | 718.6 | reference |
| Blender Benchmark 2.83GPU, victor (CUDA) | 690.6 | 708.4 | reference |
| Blender Benchmark 2.91.2GPU, victor (CUDA) | 668.2 | 684.3 | reference |
| Blender Benchmark 2.92GPU, victor (CUDA) | 745.5 | 805.1 | reference |
| Blender Benchmark 2.93.1GPU, victor (CUDA) | 734.5 | 706.2 | reference |
| Blender Benchmark 2.90GPU, victor (OPTIX) | 650.3 | 585.4 | reference |
| Blender Benchmark 2.92GPU, victor (OPTIX) | 751.1 | — | no pair |
| Blender Benchmark 2.93.1GPU, victor (OPTIX) | 701.6 | 690.1 | reference |
| Blender Benchmark 3.1.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 687.1 | 851.2 | reference |
| Blender Benchmark 3.2.1GPU, sum of scenes monster + junkshop + classroom (CUDA) | 674.9 | 883.7 | reference |
| Blender Benchmark 3.3.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 652.1 | 898.2 | reference |
| Blender Benchmark 3.4.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 667.3 | 849.6 | reference |
| Blender Benchmark 3.5.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 663.7 | 907.5 | reference |
| Blender Benchmark 3.6.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 684 | 971.3 | reference |
| Blender Benchmark 4.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 592.8 | 763.4 | reference |
| Blender Benchmark 5.1.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 495 | 702.6 | reference |
| Blender Benchmark 5.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 513.6 | 685.6 | reference |
| Blender Benchmark 3.0.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 658.3 | 888.1 | reference |
| Blender Benchmark 3.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 662.6 | 813 | reference |
| Blender Benchmark 3.2.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 614.7 | 810.8 | reference |
| Blender Benchmark 4.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 599.7 | 788.7 | reference |
| Blender Benchmark 4.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 590.9 | 765.6 | reference |
| Blender Benchmark 4.4.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 587.9 | 753.9 | reference |
| Blender Benchmark 5.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 506.8 | 698.7 | reference |
| Blender Benchmark 2.91.2GPU, classroom (OPTIX) | — | 302.3 | no pair |
| Blender Benchmark 2.91.2GPU, fishy_cat (OPTIX) | — | 148.8 | no pair |
| Blender Benchmark 2.91.2GPU, koro (OPTIX) | — | 182 | no pair |
| Blender Benchmark 2.91.2GPU, pavillon_barcelona (OPTIX) | — | 427 | no pair |
| Blender Benchmark 2.91.2GPU, victor (OPTIX) | — | 598.2 | no pair |
| Blender Benchmark 4.2.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | — | 773.7 | no pair |
| Blender Benchmark 5.2.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | — | 777.5 | no pair |
| Blender Benchmark 3.1.2GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 875.8 | no pair |
| Blender Benchmark 5.0.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 686.9 | no pair |
Each benchmark version has its own scale: numbers are compared only within a row.
Specifications
| Specification | A · GeForce GTX 1070 Ti | B · GeForce GTX 1660 Ti | Difference |
|---|---|---|---|
| Chip | |||
| GPU chip | GP104-300-A1 | TU116-400-A1 | differs |
| Architecture | Pascal | Turing | differs |
| Shader unitsShader unit counts are comparable only within one architecture. | 2,432 (CUDA cores) | 1,536 (CUDA cores) | not directly comparable |
| Clocks | |||
| Base clock | 1,607 MHz | 1,500 MHz | A: 107 MHz more |
| Boost clockClocks of different architectures do not reflect performance directly. For laptop chips, the clock depends on the laptop’s power limit. | 1,683 MHz | 1,770 MHz | B: 87 MHz more |
| Memory | |||
| Video memory | 8 GB | 6 GB | A: 2 GB more |
| Memory type | GDDR5 | GDDR6 | differs |
| Memory bus | 256 bit | 192 bit | A: 64 bit more |
| Memory bandwidthExcludes cache (Infinity Cache / L2), which affects effective bandwidth differently. | 256.3 GB/s | 288 GB/s | B: 31.7 GB/s more |
| Power and dimensions | |||
| Board powerTGP (NVIDIA) and TBP (AMD/Intel) are calculated differently; higher is not better. For laptops, the chip maker sets the range and the laptop maker picks the value, so the same chip behaves differently in different laptops. | 180 W (TDP) | 120 W (TDP) | A: 60 W higher |
| Interface | PCIe 3.0 x16 | no data | no data for B |
Chip
GPU chip
AGP104-300-A1
BTU116-400-A1
differsArchitecture
APascal
BTuring
differsShader units
A2,432 (CUDA cores)
B1,536 (CUDA cores)
not directly comparableShader unit counts are comparable only within one architecture.
Clocks
Base clock
A1,607 MHz
B1,500 MHz
A: 107 MHz moreBoost clock
A1,683 MHz
B1,770 MHz
B: 87 MHz moreClocks of different architectures do not reflect performance directly. For laptop chips, the clock depends on the laptop’s power limit.
Memory
Video memory
A8 GB
B6 GB
A: 2 GB moreMemory type
AGDDR5
BGDDR6
differsMemory bus
A256 bit
B192 bit
A: 64 bit moreMemory bandwidth
A256.3 GB/s
B288 GB/s
B: 31.7 GB/s moreExcludes cache (Infinity Cache / L2), which affects effective bandwidth differently.
Power and dimensions
Board power
A180 W (TDP)
B120 W (TDP)
A: 60 W higherTGP (NVIDIA) and TBP (AMD/Intel) are calculated differently; higher is not better. For laptops, the chip maker sets the range and the laptop maker picks the value, so the same chip behaves differently in different laptops.
Interface
APCIe 3.0 x16
Bno data
no data for B
“More” does not always mean “better”: higher power draw or size is a system requirement, not an advantage (such bars are dimmed).
Compare with AI
An AI model reviews the specs and benchmarks of this pair and gives a plain-language verdict on who each model suits.
Questions and answers
How much video memory do NVIDIA GeForce GTX 1070 Ti and NVIDIA GeForce GTX 1660 Ti have?
NVIDIA GeForce GTX 1070 Ti: 8 GB GDDR5, 256-bit bus. NVIDIA GeForce GTX 1660 Ti: 6 GB GDDR6, 192-bit bus.
Which is faster: NVIDIA GeForce GTX 1070 Ti or NVIDIA GeForce GTX 1660 Ti?
It depends on the task: rendering and compute: results disagree.
See also
Data sources (2)
- Wikipedia (English) — specifications of NVIDIA GeForce GTX 1070 Ti, specifications of NVIDIA GeForce GTX 1660 Ti, release date and launch pricechecked November 14, 2026
- Blender Open Data (Blender Foundation) — benchmark Blender Benchmark 2.81, benchmark Blender Benchmark 2.81a, benchmark Blender Benchmark 2.82, benchmark Blender Benchmark 2.83, benchmark Blender Benchmark 2.90, benchmark Blender Benchmark 2.91.2, benchmark Blender Benchmark 2.92, benchmark Blender Benchmark 2.93.1, benchmark Blender Benchmark 3.1.0, benchmark Blender Benchmark 3.2.1, benchmark Blender Benchmark 3.3.0, benchmark Blender Benchmark 3.4.0, benchmark Blender Benchmark 3.5.0, benchmark Blender Benchmark 3.6.0, benchmark Blender Benchmark 4.2.0, benchmark Blender Benchmark 5.2.0, benchmark Blender Benchmark 4.5.0, benchmark Blender Benchmark 5.1.1, benchmark Blender Benchmark 3.0.1, benchmark Blender Benchmark 3.2.0, benchmark Blender Benchmark 4.0.0, benchmark Blender Benchmark 4.1.0, benchmark Blender Benchmark 4.3.0, benchmark Blender Benchmark 4.4.0, benchmark Blender Benchmark 5.1.0, benchmark Blender Benchmark 3.1.2, benchmark Blender Benchmark 5.0.0checked November 14, 2026
How we compare: methodology. Where the data comes from: sources.