NVIDIA GeForce 825M
The NVIDIA GeForce 825M is a graphics card released on January 27, 2014, built on the Kepler architecture. It has 2 GB of DDR3 memory on a 64-bit bus, 14.4 GB/s of memory bandwidth, a boost clock of 850 MHz and a 33 W board power rating. It scores 749 in the PassMark G3D benchmark. It supports DirectX 12, OpenGL 4.5, Vulkan 1.2 and CUDA compute capability 3.5.
Depth: 34 recorded fields, 1 feature, 5 PassMark workloads
Specification data last verified October 1, 2026 against PassMark Videocard Benchmark and Wikipedia GPU list tables (wikigrid-2026-09). Coverage: 8 of 10 tracked specifications recorded.
Performance
The NVIDIA GeForce 825M scores 749 in the PassMark G3D benchmark, higher than 47.6% of the 2,658 GPUs benchmarked (PassMark rank #1,585), measured across 43 submitted results. Its PassMark G2D score is 200.
PassMark workload tests
In the PassMark workload tests the NVIDIA GeForce 825M averages DirectX 9 15 fps, DirectX 10 2 fps, DirectX 11 5 fps, DirectX 12 4 fps and GPU Compute 525 ops/s, across 43 submitted results. Frames per second are PassMark test-scene rates, not game frame rates.
As stated by PassMark
PassMark's own listing for the NVIDIA GeForce 825M states the PCIe 3.0 x8 interface, up to 2 GB of memory, a core clock of 850 MHz, DirectX 11 and OpenGL 4.5, first benchmarked on 2014-10-07. These are PassMark's figures, shown separately from the specification tables above.
Graphics processor
The NVIDIA GeForce 825M is based on the Kepler architecture and uses the GK208 graphics processor.
Clock speeds
The NVIDIA GeForce 825M runs at a base clock of 850 MHz.
Memory
The NVIDIA GeForce 825M is equipped with 2 GB of DDR3 memory, uses a 64-bit memory bus and delivers 14.4 GB/s of memory bandwidth.
Board and power
The NVIDIA GeForce 825M has a board power rating of 33 W and it connects over PCIe 3.0 x8.
Market
The NVIDIA GeForce 825M was released on January 27, 2014.
Benchmarks
Graphics and compute APIs
The NVIDIA GeForce 825M supports DirectX 12, OpenGL 4.5, Vulkan 1.2 and CUDA compute capability 3.5.
Feex has no recorded OpenCL version for the NVIDIA GeForce 825M.
Features
The NVIDIA GeForce 825M has 1 supported feature recorded outside the media engine. A feature that no source mentions is shown as not listed, never as unsupported.
Compute and AI
-
Supported: CUDA Compute capability 3.5Support for NVIDIA's CUDA computing platform, used by many creative, scientific and AI applications.
Not listed (2)
No source Feex holds states whether the NVIDIA GeForce 825M supports Tensor cores and NVLink / SLI. That is silence, not a statement that they are unsupported.
The NVIDIA GeForce 825M supports CUDA (Compute capability 3.5) under Compute and AI.
Naming and aliases
The NVIDIA GeForce 825M is also known by other names: PassMark also lists it as NVIDIA GeForce 825M (Mobile).
Frequently asked questions
How much VRAM does the NVIDIA GeForce 825M have?
The NVIDIA GeForce 825M has 2 GB of DDR3 memory on a 64-bit bus. Its memory bandwidth is 14.4 GB/s.
How much power does the NVIDIA GeForce 825M use and what power supply does it need?
The NVIDIA GeForce 825M has a board power rating of 33 W.
How fast is the NVIDIA GeForce 825M?
The NVIDIA GeForce 825M scores 749 in the PassMark G3D benchmark, higher than 47.6% of the 2,658 GPUs benchmarked. Its PassMark rank is #1,585. In the PassMark DirectX 12 test it averages 4 fps.
What is the CUDA compute capability of the NVIDIA GeForce 825M?
The NVIDIA GeForce 825M has CUDA compute capability 3.5.
What bus interface does the NVIDIA GeForce 825M use?
The NVIDIA GeForce 825M connects over PCIe 3.0 x8.
When was the NVIDIA GeForce 825M released?
The NVIDIA GeForce 825M was released on January 27, 2014.
Corrections to this record
After an audit, 1 empty field was filled from sources. Each change is listed with its old value, new value, rule and sources.
1 filled value
- Architecture: Kepler (was Not recorded). Rule: Architecture from chip codename. chip GK208 -> Kepler. Sources: Wikipedia: List of Nvidia graphics processing units (revision 1377107724).
The audit log is public: all changes for this GPU, the corrections overview and the method.