GeForce RTX 3070 Ti vs GeForce RTX 3080: specs and benchmark comparison
Short verdict
NVIDIA GeForce RTX 3080 leads in every scenario with comparable benchmarks: rendering and compute. This does not imply an advantage in tasks without benchmarks.
Compare with AIRendering and compute
B +39.4%NVIDIA GeForce RTX 3080 leads in this scenario (up to 39.4% in comparable benchmarks).
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 RTX 3070 Ti is better1
Board power 30 W lower290 W (board power) vs 320 W (board power)
Where GeForce RTX 3080 is better16
Blender Benchmark 2.93.1 27.3% lower20 s vs 25 s
Blender Benchmark 2.93.1 30.6% lower11 s vs 15 s
Blender Benchmark 2.93.1 17.9% lower94 s vs 111 s
Blender Benchmark 2.93.1 24.6% lower40 s vs 50 s
Blender Benchmark 2.93.1 32.8% lower40 s vs 53 s
Blender Benchmark 2.93.1 38% lower22 s vs 30 s
Blender Benchmark 2.93.1 36.9% lower65 s vs 88 s
Blender Benchmark 2.93.1 39.4% lower45 s vs 63 s
Blender Benchmark 2.93.1 25% lower119 s vs 149 s
Blender Benchmark 2.93.1 26.2% lower54 s vs 68 s
Blender Benchmark 2.93.1 29.5% lower166 s vs 215 s
Blender Benchmark 2.93.1 30.2% lower72 s vs 93 s
Blender Benchmark 4.5.0 28.8% faster4,527 samples/min vs 3,514 samples/min
Shader units 2,560 more8,704 (CUDA cores) vs 6,144 (CUDA cores)
Video memory 2 GB more10 GB vs 8 GB
Memory bus 64 bit more320 bit vs 256 bit
Benchmarks
Results are compared within one benchmark, one version and the same conditions.
- B ahead by 27.3%
Blender Benchmark 2.93.1
GPU, bmw27 (CUDA) · s, lower is better
GeForce RTX 3070 Ti25.3 sGeForce RTX 308019.9 sPosition among tested models (31, CUDA)
B · 20A · 250100200300Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 27.3% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
25.3 s · community measurement, sample 90 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 90 submissions (19 outliers excluded), interquartile range 25.14–25.68; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
19.9 s · community measurement, sample 293 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 293 submissions (37 outliers excluded), interquartile range 19.64–20.28; submissions 2021-07-09 to 2022-03-01; 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 30.6%
Blender Benchmark 2.93.1
GPU, bmw27 (OPTIX) · s, lower is better
GeForce RTX 3070 Ti14.8 sGeForce RTX 308011.4 sPosition among tested models (20, OptiX)
B · 11A · 15050100150Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 30.6% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
14.8 s · community measurement, sample 42 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 42 submissions (7 outliers excluded), interquartile range 14.72–15.04; submissions 2021-07-10 to 2022-02-21; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
11.4 s · community measurement, sample 114 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 114 submissions (36 outliers excluded), interquartile range 11.26–11.53; submissions 2021-07-09 to 2022-02-28; OPTIX 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 17.9%
Blender Benchmark 2.93.1
GPU, classroom (CUDA) · s, lower is better
GeForce RTX 3070 Ti110.6 sGeForce RTX 308093.8 sPosition among tested models (26, CUDA)
B · 94A · 1110200400600800Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 17.9% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
110.6 s · community measurement, sample 73 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 73 submissions (15 outliers excluded), interquartile range 108.88–111.47; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
93.8 s · community measurement, sample 241 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 241 submissions (35 outliers excluded), interquartile range 92.51–95.41; submissions 2021-07-09 to 2022-03-01; 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 24.6%
Blender Benchmark 2.93.1
GPU, classroom (OPTIX) · s, lower is better
GeForce RTX 3070 Ti49.8 sGeForce RTX 308039.9 sPosition among tested models (14, OptiX)
B · 40A · 50050100150200Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 24.6% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
49.8 s · community measurement, sample 42 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 42 submissions (4 outliers excluded), interquartile range 49.1–50.9; submissions 2021-07-10 to 2022-02-21; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
39.9 s · community measurement, sample 136 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 136 submissions (6 outliers excluded), interquartile range 39.04–40.87; submissions 2021-07-09 to 2022-02-28; OPTIX 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 32.8%
Blender Benchmark 2.93.1
GPU, fishy_cat (CUDA) · s, lower is better
GeForce RTX 3070 Ti53.4 sGeForce RTX 308040.2 sPosition among tested models (21, CUDA)
B · 40A · 530100200300Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 32.8% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
53.4 s · community measurement, sample 65 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 65 submissions (10 outliers excluded), interquartile range 52.81–54.09; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
40.2 s · community measurement, sample 210 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 210 submissions (41 outliers excluded), interquartile range 39.77–40.76; submissions 2021-07-09 to 2022-03-01; 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 38%
Blender Benchmark 2.93.1
GPU, fishy_cat (OPTIX) · s, lower is better
GeForce RTX 3070 Ti29.9 sGeForce RTX 308021.6 sPosition among tested models (13, OptiX)
B · 22A · 30020406080Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 38% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
29.9 s · community measurement, sample 32 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 32 submissions (3 outliers excluded), interquartile range 29.51–30.21; submissions 2021-07-10 to 2022-02-14; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
21.6 s · community measurement, sample 87 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 87 submissions (33 outliers excluded), interquartile range 21.48–21.83; submissions 2021-07-09 to 2022-02-28; OPTIX 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 36.9%
Blender Benchmark 2.93.1
GPU, koro (CUDA) · s, lower is better
GeForce RTX 3070 Ti88.4 sGeForce RTX 308064.6 sPosition among tested models (20, CUDA)
B · 65A · 880200400600Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 36.9% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
88.4 s · community measurement, sample 60 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 60 submissions (11 outliers excluded), interquartile range 87.85–89.37; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
64.6 s · community measurement, sample 212 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 212 submissions (34 outliers excluded), interquartile range 63.85–65.6; submissions 2021-07-09 to 2022-03-01; 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 39.4%
Blender Benchmark 2.93.1
GPU, koro (OPTIX) · s, lower is better
GeForce RTX 3070 Ti63.2 sGeForce RTX 308045.3 sPosition among tested models (13, OptiX)
B · 45A · 63050100150200Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 39.4% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
63.2 s · community measurement, sample 34 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 34 submissions (0 outliers excluded), interquartile range 62.36–64.49; submissions 2021-07-10 to 2022-02-14; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
45.3 s · community measurement, sample 103 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 103 submissions (13 outliers excluded), interquartile range 44.05–45.87; submissions 2021-07-09 to 2022-02-28; OPTIX 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 25%
Blender Benchmark 2.93.1
GPU, pavillon_barcelona (CUDA) · s, lower is better
GeForce RTX 3070 Ti148.7 sGeForce RTX 3080119 sPosition among tested models (21, CUDA)
B · 119A · 1490200400600800Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 25% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
148.7 s · community measurement, sample 60 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 60 submissions (12 outliers excluded), interquartile range 147.41–150.7; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
119 s · community measurement, sample 211 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 211 submissions (34 outliers excluded), interquartile range 117.52–120.77; submissions 2021-07-09 to 2022-03-01; 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 26.2%
Blender Benchmark 2.93.1
GPU, pavillon_barcelona (OPTIX) · s, lower is better
GeForce RTX 3070 Ti68.4 sGeForce RTX 308054.2 sPosition among tested models (13, OptiX)
B · 54A · 680100200300Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 26.2% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
68.4 s · community measurement, sample 37 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 37 submissions (3 outliers excluded), interquartile range 67.81–69.84; submissions 2021-07-10 to 2022-02-21; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
54.2 s · community measurement, sample 102 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 102 submissions (19 outliers excluded), interquartile range 53.49–54.93; submissions 2021-07-09 to 2022-02-28; OPTIX 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 29.5%
Blender Benchmark 2.93.1
GPU, victor (CUDA) · s, lower is better
GeForce RTX 3070 Ti214.7 sGeForce RTX 3080165.8 sPosition among tested models (19, CUDA)
B · 166A · 21502505007501,000Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 29.5% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
214.7 s · community measurement, sample 52 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 52 submissions (15 outliers excluded), interquartile range 213.55–217.22; submissions 2021-07-12 to 2022-02-27; CUDA backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
165.8 s · community measurement, sample 205 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 205 submissions (31 outliers excluded), interquartile range 164.1–168.46; submissions 2021-07-09 to 2022-03-01; 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 30.2%
Blender Benchmark 2.93.1
GPU, victor (OPTIX) · s, lower is better
GeForce RTX 3070 Ti93.4 sGeForce RTX 308071.7 sPosition among tested models (13, OptiX)
B · 72A · 930100200300Each dot is one model’s median in this benchmark. The scale starts at zero, s. NVIDIA GeForce RTX 3080 is ahead by 30.2% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
93.4 s · community measurement, sample 35 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 35 submissions (0 outliers excluded), interquartile range 92.93–96.09; submissions 2021-07-10 to 2022-02-21; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
71.7 s · community measurement, sample 109 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 109 submissions (14 outliers excluded), interquartile range 70.08–72.69; submissions 2021-07-09 to 2022-02-28; OPTIX 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 28.8%
Blender Benchmark 4.5.0
GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) · samples/min, higher is better
GeForce RTX 3070 Ti3,514.2 samples/minGeForce RTX 30804,526.6 samples/minPosition among tested models (67, OptiX)
A · 3,514B · 4,52705,00010,00015,000Each dot is one model’s median in this benchmark. The scale starts at zero, samples/min. NVIDIA GeForce RTX 3080 is ahead by 28.8% in this benchmark.
About the measurement and source
- A · NVIDIA GeForce RTX 3070 Ti
3,514.2 samples/min · community measurement, sample 109 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 109 submissions (9 outliers excluded), interquartile range 3371.91–3561.39; submissions 2025-07-17 to 2026-11-11; OPTIX backend. Snapshot 2026-11-14.
- B · NVIDIA GeForce RTX 3080
4,526.6 samples/min · community measurement, sample 536 · November 14, 2026
Blender Open Data (Blender Foundation)
Median of 536 submissions (43 outliers excluded), interquartile range 4378.6–4702.09; submissions 2025-07-17 to 2026-09-13; 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 (15)
| Benchmark | A | B | Difference |
|---|---|---|---|
| Blender Benchmark 4.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,488.7 | 4,450.1 | B +27.6% |
| Blender Benchmark 5.1.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,286.8 | 4,379.9 | B +33.3% |
| Blender Benchmark 5.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,395.4 | 4,197.4 | B +23.6% |
| Blender Benchmark 3.0.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,849.2 | 5,065.1 | B +31.6% |
| Blender Benchmark 3.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,793 | 4,893.1 | B +29% |
| Blender Benchmark 3.2.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,805.6 | 4,851.8 | B +27.5% |
| Blender Benchmark 3.2.1GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,795.9 | 4,870.7 | B +28.3% |
| Blender Benchmark 3.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,888.7 | 5,025.1 | B +29.2% |
| Blender Benchmark 3.4.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,891.1 | 5,001.5 | B +28.5% |
| Blender Benchmark 3.5.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 4,130.4 | 5,251.5 | B +27.1% |
| Blender Benchmark 3.6.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 4,100.2 | 5,201.5 | B +26.9% |
| Blender Benchmark 4.0.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,599.4 | 4,564.4 | B +26.8% |
| Blender Benchmark 4.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,624.6 | 4,614.7 | B +27.3% |
| Blender Benchmark 4.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,495.5 | 4,511.6 | B +29.1% |
| Blender Benchmark 4.4.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,501.2 | 4,478.6 | B +27.9% |
Each benchmark version has its own scale: numbers are compared only within a row.
Reference results (53)
| Benchmark | A | B | Difference |
|---|---|---|---|
| Blender Benchmark 2.91.2GPU, bmw27 (CUDA) | 28.5 | 21.7 | reference |
| Blender Benchmark 2.92GPU, bmw27 (CUDA) | 27.8 | 21.8 | reference |
| Blender Benchmark 2.91.2GPU, bmw27 (OPTIX) | 13.7 | 11 | reference |
| Blender Benchmark 2.92GPU, bmw27 (OPTIX) | 14 | 10.7 | reference |
| Blender Benchmark 2.91.2GPU, classroom (CUDA) | 81.9 | 68.6 | reference |
| Blender Benchmark 2.92GPU, classroom (CUDA) | 79.3 | 65.3 | reference |
| Blender Benchmark 2.91.2GPU, classroom (OPTIX) | 48.1 | 42.5 | reference |
| Blender Benchmark 2.92GPU, classroom (OPTIX) | 48.9 | 39.5 | reference |
| Blender Benchmark 2.91.2GPU, fishy_cat (CUDA) | 52.8 | 39.5 | reference |
| Blender Benchmark 2.92GPU, fishy_cat (CUDA) | 53.8 | 40.8 | reference |
| Blender Benchmark 2.91.2GPU, fishy_cat (OPTIX) | 27 | 19.7 | reference |
| Blender Benchmark 2.92GPU, fishy_cat (OPTIX) | 29.4 | 21.4 | reference |
| Blender Benchmark 2.91.2GPU, koro (CUDA) | 92 | 68.4 | reference |
| Blender Benchmark 2.92GPU, koro (CUDA) | 92.1 | 68.9 | reference |
| Blender Benchmark 2.91.2GPU, koro (OPTIX) | 60.5 | 42.1 | reference |
| Blender Benchmark 2.92GPU, koro (OPTIX) | 64.4 | 45.8 | reference |
| Blender Benchmark 2.91.2GPU, pavillon_barcelona (CUDA) | 164.2 | 129.7 | reference |
| Blender Benchmark 2.92GPU, pavillon_barcelona (CUDA) | 164.4 | 130 | reference |
| Blender Benchmark 2.91.2GPU, pavillon_barcelona (OPTIX) | 66.6 | 54.5 | reference |
| Blender Benchmark 2.92GPU, pavillon_barcelona (OPTIX) | 65.9 | 53 | reference |
| Blender Benchmark 2.91.2GPU, victor (CUDA) | 195.7 | 150.7 | reference |
| Blender Benchmark 2.92GPU, victor (CUDA) | 259.7 | 207.5 | reference |
| Blender Benchmark 2.91.2GPU, victor (OPTIX) | 77.7 | 62.7 | reference |
| Blender Benchmark 2.92GPU, victor (OPTIX) | 93.5 | 70.9 | reference |
| Blender Benchmark 3.1.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 2,299.5 | 2,981.3 | reference |
| Blender Benchmark 3.2.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 2,102.2 | 2,845.2 | reference |
| Blender Benchmark 3.3.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 2,039.7 | 2,902.8 | reference |
| Blender Benchmark 3.6.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 2,605 | 3,193.9 | reference |
| Blender Benchmark 4.2.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | 2,339.1 | — | no pair |
| Blender Benchmark 5.0.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,076.6 | 4,105.5 | reference |
| Blender Benchmark 5.1.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,325.5 | 4,330.8 | reference |
| Blender Benchmark 5.3.0GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | 3,665.8 | 4,546.6 | reference |
| Blender Benchmark 2.90GPU, bmw27 (CUDA) | — | 25.5 | no pair |
| Blender Benchmark 2.90.1GPU, bmw27 (CUDA) | — | 25.4 | no pair |
| Blender Benchmark 2.90GPU, bmw27 (OPTIX) | — | 10.7 | no pair |
| Blender Benchmark 2.90GPU, classroom (CUDA) | — | 78.1 | no pair |
| Blender Benchmark 2.90.1GPU, classroom (CUDA) | — | 96.6 | no pair |
| Blender Benchmark 2.90GPU, classroom (OPTIX) | — | 41.4 | no pair |
| Blender Benchmark 2.90GPU, fishy_cat (CUDA) | — | 45.8 | no pair |
| Blender Benchmark 2.90.1GPU, fishy_cat (CUDA) | — | 46.4 | no pair |
| Blender Benchmark 2.90GPU, fishy_cat (OPTIX) | — | 19.8 | no pair |
| Blender Benchmark 2.90GPU, koro (CUDA) | — | 68.4 | no pair |
| Blender Benchmark 2.90.1GPU, koro (CUDA) | — | 68.4 | no pair |
| Blender Benchmark 2.90GPU, koro (OPTIX) | — | 42.7 | no pair |
| Blender Benchmark 2.90GPU, pavillon_barcelona (CUDA) | — | 145.3 | no pair |
| Blender Benchmark 2.90.1GPU, pavillon_barcelona (CUDA) | — | 139.1 | no pair |
| Blender Benchmark 2.90GPU, pavillon_barcelona (OPTIX) | — | 53.6 | no pair |
| Blender Benchmark 2.90GPU, victor (CUDA) | — | 191.6 | no pair |
| Blender Benchmark 2.90.1GPU, victor (CUDA) | — | 195 | no pair |
| Blender Benchmark 2.90GPU, victor (OPTIX) | — | 60.7 | no pair |
| Blender Benchmark 3.5.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | — | 3,291.3 | no pair |
| Blender Benchmark 4.0.0GPU, sum of scenes monster + junkshop + classroom (CUDA) | — | 2,989.8 | no pair |
| Blender Benchmark 3.1.2GPU, sum of scenes monster + junkshop + classroom (OptiX / HIP / oneAPI / Metal) | — | 5,100.4 | no pair |
Each benchmark version has its own scale: numbers are compared only within a row.
Specifications
| Specification | A · GeForce RTX 3070 Ti | B · GeForce RTX 3080 | Difference |
|---|---|---|---|
| Chip | |||
| Architecture | Ampere | Ampere | equal |
| Shader unitsShader unit counts are comparable only within one architecture. | 6,144 (CUDA cores) | 8,704 (CUDA cores) | B: 2,560 more |
| Clocks | |||
| Base clock | 1,580 MHz | 1,440 MHz | A: 140 MHz more |
| Boost clockClocks of different architectures do not reflect performance directly. For laptop chips, the clock depends on the laptop’s power limit. | 1,770 MHz | 1,710 MHz | A: 60 MHz more |
| Memory | |||
| Video memory | 8 GB | 10 GB | B: 2 GB more |
| Memory type | GDDR6X | GDDR6X | equal |
| Memory bus | 256 bit | 320 bit | B: 64 bit 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. | 290 W (board power) | 320 W (board power) | B: 30 W higher |
| Recommended PSU | 750 W | 750 W | equal |
| Interface | no data | PCIe 4.0 | no data for A |
| Length (reference / Founders Edition)Partner cards have different dimensions; check the specific SKU. | 267 mm | 285 mm | B: 18 mm higher |
| Thickness (slots) | 2 | 2 | equal |
| Power connectors (reference) | 2x 8-pin (adapter to 1x 12-pin included) — FE | 2x 8-pin (adapter to 1x 12-pin included) — Founders Edition | differs |
Chip
Architecture
AAmpere
BAmpere
equalShader units
A6,144 (CUDA cores)
B8,704 (CUDA cores)
B: 2,560 moreShader unit counts are comparable only within one architecture.
Clocks
Base clock
A1,580 MHz
B1,440 MHz
A: 140 MHz moreBoost clock
A1,770 MHz
B1,710 MHz
A: 60 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
B10 GB
B: 2 GB moreMemory type
AGDDR6X
BGDDR6X
equalMemory bus
A256 bit
B320 bit
B: 64 bit more
Power and dimensions
Board power
A290 W (board power)
B320 W (board power)
B: 30 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.
Recommended PSU
A750 W
B750 W
equalInterface
Ano data
BPCIe 4.0
no data for ALength (reference / Founders Edition)
A267 mm
B285 mm
B: 18 mm higherPartner cards have different dimensions; check the specific SKU.
Thickness (slots)
A2
B2
equalPower connectors (reference)
A2x 8-pin (adapter to 1x 12-pin included) — FE
B2x 8-pin (adapter to 1x 12-pin included) — Founders Edition
differs
“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 RTX 3070 Ti and NVIDIA GeForce RTX 3080 have?
NVIDIA GeForce RTX 3070 Ti: 8 GB GDDR6X, 256-bit bus. NVIDIA GeForce RTX 3080: 10 GB GDDR6X, 320-bit bus.
What power supply does the manufacturer recommend?
For NVIDIA GeForce RTX 3070 Ti, 750 W or more; for NVIDIA GeForce RTX 3080, 750 W or more. This is a recommendation for a typical system; also check the power connectors of the specific card model.
Which is faster: NVIDIA GeForce RTX 3070 Ti or NVIDIA GeForce RTX 3080?
NVIDIA GeForce RTX 3080 leads in every scenario with comparable benchmarks: rendering and compute. This does not imply an advantage in tasks without benchmarks.
See also
Data sources (3)
- NVIDIA — specifications of NVIDIA GeForce RTX 3070 Ti, specifications of NVIDIA GeForce RTX 3080checked November 7, 2026
- Blender Open Data (Blender Foundation) — benchmark Blender Benchmark 2.91.2, benchmark Blender Benchmark 2.92, benchmark Blender Benchmark 2.93.1, benchmark Blender Benchmark 2.90, benchmark Blender Benchmark 2.90.1, benchmark Blender Benchmark 3.1.0, benchmark Blender Benchmark 3.2.0, benchmark Blender Benchmark 3.3.0, benchmark Blender Benchmark 3.6.0, benchmark Blender Benchmark 4.2.0, benchmark Blender Benchmark 3.5.0, benchmark Blender Benchmark 4.0.0, benchmark Blender Benchmark 4.5.0, benchmark Blender Benchmark 5.1.1, benchmark Blender Benchmark 5.2.0, benchmark Blender Benchmark 3.0.1, benchmark Blender Benchmark 3.2.1, benchmark Blender Benchmark 3.4.0, benchmark Blender Benchmark 4.1.0, benchmark Blender Benchmark 4.3.0, benchmark Blender Benchmark 4.4.0, benchmark Blender Benchmark 5.0.0, benchmark Blender Benchmark 5.1.0, benchmark Blender Benchmark 5.3.0, benchmark Blender Benchmark 3.1.2checked November 14, 2026
- Wikipedia (English) — release date and launch pricechecked November 14, 2026
How we compare: methodology. Where the data comes from: sources.