910M 2 GB for Blender
Kepler 2.0 GPU with 2 GB VRAM, below Blender’s recommended 8 GB. Check scene memory needs and GPU compatibility before buying.
Last updated: September 23, 2026
Compare 910M 2 GB to...
Pick another GPU to open a side-by-side comparison.
Price unavailable
No current offer matches your preferences.
- Blender Cycles score
- Unavailable
- VRAM
- 2 GB
- Below Blender’s recommended VRAM
- Released
- 2015
- Render support
- CUDA, OpenCL
- Power draw (TDP)
- 33 W
Technical specifications
GPU
- ArchitectureKepler 2.0
- CUDA cores384
Memory
- Bandwidth16 GB/s
- Bus width64-bit
- Memory typeDDR3
Clocks & manufacturing
- Boost clock641 MHz
- Base clock641 MHz
- Process size28 nm
- GPU chipGK208B
Power & PSU sources
GPU power ratings
TDP: 33 W
Secondary source: RightNow GPU Database / TechPowerUp ↗ · Checked Sep 23, 2026 · Reference card
GPU power ratings and whole-system PSU recommendations measure different things. Check the exact card and the source’s system assumptions.
910M 2 GB Blender benchmarks
Estimated time to render one benchmark frame · Lower is faster.
Scene benchmarks unavailable
Scene benchmarks are unavailable for this GPU’s selected benchmark result.
About these estimates
Times estimate one frame of each benchmark scene from the same runs, Blender version and backend as the displayed Cycles score. The calculation uses 240 samples for Junkshop, 256 for Monster and 300 for Classroom, plus a 19% estimation buffer. Actual project render times vary. Results from different Blender versions may not be directly comparable.
Is 910M 2 GB good for Blender?
A concise editorial read on where this GPU looks strong, the tradeoffs to keep in mind, and who it suits best.
Tradeoffs to know
- Meets Blender’s 2 GB VRAM minimum but falls below its 8 GB recommendation for Windows/Linux. Usability depends on GPU and driver compatibility and your scene’s memory needs.
Who should choose it
- Kepler 2.0 GPU with 2 GB VRAM, below Blender’s recommended 8 GB. Check scene memory needs and GPU compatibility before buying.