NVIDIA · Maxwell
NVIDIA GeForce GTX 970M: specs, benchmarks and comparison
- 75 W
- Release
- October 7, 2014
- Power
- 75 W
- depends on the laptop model
Performance
Rank among models tested in the same benchmark on the same render backend. A laptop chip’s result depends on the laptop model: its maker sets the power limit and cooling.
Blender Benchmark 2.93.1
GPU, bmw27 (CUDA)
180.13 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (5)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.92 | 274.38 s · small sample | — |
| Blender Benchmark 2.90 | 259.59 s · small sample | — |
| Blender Benchmark 2.83 | 236.79 s · small sample | — |
| Blender Benchmark 2.82 | 226.54 s · small sample | — |
| Blender Benchmark 2.81a | 234.18 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 2.92
GPU, classroom (CUDA)
909.59 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (4)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.91.2 | 770.94 s · small sample | — |
| Blender Benchmark 2.90 | 865.04 s · small sample | — |
| Blender Benchmark 2.83 | 834.02 s · small sample | — |
| Blender Benchmark 2.82 | 793.5 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 2.92
GPU, fishy_cat (CUDA)
722.47 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (4)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.90 | 503.83 s · small sample | — |
| Blender Benchmark 2.83 | 514.54 s · small sample | — |
| Blender Benchmark 2.82 | 498.87 s · small sample | — |
| Blender Benchmark 2.81a | 485.48 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 2.90
GPU, koro (CUDA)
890.79 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (2)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.83 | 1,054.79 s · small sample | — |
| Blender Benchmark 2.82 | 1,032.68 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 2.90
GPU, pavillon_barcelona (CUDA)
1,264.88 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (2)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.83 | 1,305.73 s · small sample | — |
| Blender Benchmark 2.82 | 1,242.01 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 2.90
GPU, victor (CUDA)
2,588.66 s
lower is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (2)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 2.82 | 2,301.38 s · small sample | — |
| Blender Benchmark 2.81a | 0.56 s · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Blender Benchmark 4.3.0
GPU, sum of scenes monster + junkshop + classroom (CUDA)
157 samples/min
higher is better
Median of 1 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (7)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 4.0.0 | 270 samples/min · small sample | — |
| Blender Benchmark 3.6.0 | 176 samples/min · small sample | — |
| Blender Benchmark 3.5.0 | 278 samples/min · small sample | — |
| Blender Benchmark 3.4.0 | 256 samples/min · small sample | — |
| Blender Benchmark 3.3.0 | 265 samples/min · small sample | — |
| Blender Benchmark 3.2.0 | 272 samples/min · small sample | — |
| Blender Benchmark 3.1.0 | 278 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)
184 samples/min
higher is better
Median of 2 runs · small sample · Blender Open Data (Blender Foundation) · November 14, 2026
Other benchmark versions (4)
| Benchmark | Result | Rank |
|---|---|---|
| Blender Benchmark 4.3.0 | 192 samples/min · small sample | — |
| Blender Benchmark 4.0.0 | 207 samples/min · small sample | — |
| Blender Benchmark 3.1.0 | 163 samples/min · small sample | — |
| Blender Benchmark 3.0.1 | 224 samples/min · small sample | — |
Each benchmark version has its own scale; results from different versions are not compared.
Specifications
Chip maker values. The laptop maker sets the power limit, clocks and memory in each laptop.
Chip
- GPU chip
- GM204
- Architecture
- Maxwell
Memory
- Memory type
- GDDR5
- Memory bus
- 192 bit
- Memory bandwidth
- 120 GB/s
Power and dimensions
- Board power
- 75 W (TDP)
- Interface
- PCIe 3.0 x16
Частоты and лимит мощности зависят от модели ноутбука.
Specifications: Wikipedia (English), checked November 14, 2026. Release date: Wikipedia (CC BY-SA 4.0).