Skip to content
RU

NVIDIA · Ampere

NVIDIA GeForce RTX 3080: specs, benchmarks and comparison

  • 10 GB GDDR6X
  • 320 W
  • PSU 750 W+
Manufacturer site
Illustration, not a product photo
Render rank (OptiX)
21 of 67
among models tested in Blender Benchmark 4.5.0 · 30% of the leader
Release
September 17, 2020
Power supply
750 W+
manufacturer recommendation

Performance

Rank among models tested in the same benchmark on the same render backend.

Blender Benchmark 2.93.1

GPU, bmw27 (CUDA)

19.85 s

lower is better

Rank 3 of 31 tested on CUDA85% of the leader
20
0100200300
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 293 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.9221.75 s2 of 19
Blender Benchmark 2.91.221.68 s2 of 9
Blender Benchmark 2.90.125.35 s · small sample—
Blender Benchmark 2.9025.49 s2 of 27

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, bmw27 (OPTIX)

11.35 s

lower is better

Rank 3 of 20 tested on OptiX82% of the leader
11
050100150
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 114 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9210.73 s3 of 12
Blender Benchmark 2.91.211.04 s2 of 4
Blender Benchmark 2.9010.75 s2 of 17

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, classroom (CUDA)

93.78 s

lower is better

Rank 3 of 26 tested on CUDA85% of the leader
94
0200400600800
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 241 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.9265.32 s3 of 16
Blender Benchmark 2.91.268.61 s2 of 5
Blender Benchmark 2.90.196.58 s · small sample—
Blender Benchmark 2.9078.05 s2 of 25

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, classroom (OPTIX)

39.95 s

lower is better

Rank 3 of 14 tested on OptiX84% of the leader
40
050100150200
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 136 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9239.54 s2 of 10
Blender Benchmark 2.91.242.54 s2 of 3
Blender Benchmark 2.9041.41 s2 of 15

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, fishy_cat (CUDA)

40.18 s

lower is better

Rank 3 of 21 tested on CUDA86% of the leader
40
0100200300
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 210 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.9240.83 s2 of 13
Blender Benchmark 2.91.239.46 s2 of 5
Blender Benchmark 2.90.146.42 s · small sample—
Blender Benchmark 2.9045.76 s2 of 22

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, fishy_cat (OPTIX)

21.65 s

lower is better

Rank 3 of 13 tested on OptiX83% of the leader
22
020406080
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 87 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9221.44 s2 of 9
Blender Benchmark 2.91.219.7 s2 of 3
Blender Benchmark 2.9019.84 s2 of 13

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, koro (CUDA)

64.57 s

lower is better

Rank 3 of 20 tested on CUDA85% of the leader
65
0200400600
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 212 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.9268.9 s2 of 12
Blender Benchmark 2.91.268.45 s2 of 5
Blender Benchmark 2.90.168.4 s · small sample—
Blender Benchmark 2.9068.4 s2 of 22

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, koro (OPTIX)

45.35 s

lower is better

Rank 3 of 13 tested on OptiX84% of the leader
45
050100150200
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 103 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9245.85 s1 of 10
Blender Benchmark 2.91.242.12 s2 of 3
Blender Benchmark 2.9042.66 s2 of 13

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, pavillon_barcelona (CUDA)

118.98 s

lower is better

Rank 3 of 21 tested on CUDA83% of the leader
119
0200400600800
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 211 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.92129.97 s2 of 12
Blender Benchmark 2.91.2129.72 s2 of 4
Blender Benchmark 2.90.1139.15 s · small sample—
Blender Benchmark 2.90145.31 s2 of 23

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, pavillon_barcelona (OPTIX)

54.19 s

lower is better

Rank 3 of 13 tested on OptiX83% of the leader
54
0100200300
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 102 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9252.96 s2 of 9
Blender Benchmark 2.91.254.53 s2 of 3
Blender Benchmark 2.9053.57 s2 of 13

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, victor (CUDA)

165.84 s

lower is better

Rank 3 of 19 tested on CUDA84% of the leader
166
02505007501,000
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 205 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (4)
BenchmarkResultRank
Blender Benchmark 2.92207.46 s2 of 12
Blender Benchmark 2.91.2150.73 s2 of 4
Blender Benchmark 2.90.1194.99 s · small sample—
Blender Benchmark 2.90191.56 s2 of 20

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 2.93.1

GPU, victor (OPTIX)

71.73 s

lower is better

Rank 3 of 13 tested on OptiX84% of the leader
72
0100200300
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 109 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (3)
BenchmarkResultRank
Blender Benchmark 2.9270.88 s2 of 8
Blender Benchmark 2.91.262.73 s2 of 3
Blender Benchmark 2.9060.65 s2 of 13

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 4.0.0

GPU, sum of scenes monster + junkshop + classroom (CUDA)

2,990 samples/min

higher is better

Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (5)
BenchmarkResultRank
Blender Benchmark 3.6.03,194 samples/min · small sample—
Blender Benchmark 3.5.03,291 samples/min · small sample—
Blender Benchmark 3.3.02,903 samples/min · small sample—
Blender Benchmark 3.2.02,845 samples/min · small sample—
Blender Benchmark 3.1.02,981 samples/min · small sample—

Each benchmark version has its own scale; results from different versions are not compared.

Blender Benchmark 4.5.0

GPU, sum of scenes monster + junkshop + classroom (OptiX)

4,527 samples/min

higher is better

Rank 21 of 67 tested on OptiX30% of the leader
4,527
05,00010,00015,000
Each dot is one catalog model in this benchmark; GeForce RTX 3080 is highlighted.

Median of 536 runs · Blender Open Data (Blender Foundation) · November 14, 2026

Other benchmark versions (19)
BenchmarkResultRank
Blender Benchmark 5.3.04,547 samples/min · small sample—
Blender Benchmark 5.2.04,197 samples/min19 of 48
Blender Benchmark 5.1.14,380 samples/min18 of 46
Blender Benchmark 5.1.04,331 samples/min14 of 25
Blender Benchmark 5.0.04,106 samples/min · small sample—
Blender Benchmark 4.4.04,479 samples/min20 of 59
Blender Benchmark 4.3.04,512 samples/min17 of 50
Blender Benchmark 4.2.04,450 samples/min14 of 47
Blender Benchmark 4.1.04,615 samples/min13 of 45
Blender Benchmark 4.0.04,564 samples/min13 of 47
Blender Benchmark 3.6.05,201 samples/min13 of 49
Blender Benchmark 3.5.05,252 samples/min10 of 41
Blender Benchmark 3.4.05,002 samples/min10 of 43
Blender Benchmark 3.3.05,025 samples/min8 of 37
Blender Benchmark 3.2.14,871 samples/min5 of 30
Blender Benchmark 3.2.04,852 samples/min3 of 9
Blender Benchmark 3.1.25,100 samples/min · small sample—
Blender Benchmark 3.1.04,893 samples/min4 of 38
Blender Benchmark 3.0.15,065 samples/min3 of 9

Each benchmark version has its own scale; results from different versions are not compared.

Specifications

Reference values for the chip. Clocks, length and connectors depend on the card model.

Chip

Architecture
Ampere
Shader units
8,704 (CUDA cores)

Clocks

Base clock
1,440 MHz
Boost clock
1,710 MHz

Memory

Video memory
10 GB
Memory type
GDDR6X
Memory bus
320 bit

Power and dimensions

Board power
320 W (board power)
Recommended PSU
750 W
Interface
PCIe 4.0
Length (reference / Founders Edition)
285 mm
Thickness (slots)
2
Power connectors (reference)
2x 8-pin (adapter to 1x 12-pin included) — Founders Edition

Checked against NVIDIA’s official specifications on 2026-11-14. Fields NVIDIA does not list are left empty. The original 10 GB version is listed. Blender Open Data does not include memory size in the shared name, so results apply to all variants of this model.

Specifications: NVIDIA, checked November 14, 2026. Release date: Wikipedia (CC BY-SA 4.0).

Will it fit your system

Checks the power supply, case length and PCIe slot.

  • Needs checkingPower supply. Manufacturer recommendation: 750 W or more. Enter your PSU wattage to check.
  • Needs checkingCard length and case. Enter the maximum graphics card length for your case.
  • Needs checkingPCIe interface. Enter the PCIe version of the motherboard slot.