NVIDIA
NVIDIA RTX A6000
The RTX A6000 is NVIDIA's Ampere workstation flagship, with 10,752 CUDA cores and 48 GB of GDDR6 on a 384-bit bus. It targets rendering, simulation and AI work, with NVLink support to pair two cards for 96 GB.
Pros
- 48 GB of GDDR6 memory with ECC
- 10,752 CUDA cores, 84 RT cores and 336 Tensor cores
- NVLink support to combine two cards for 96 GB
- 300 W board power with a single 8-pin connector
Cons
- Workstation pricing, with no official launch price published at announcement
- GDDR6 memory gives 768 GB/s, less than GDDR6X gaming cards
- Older Ampere design, superseded by newer workstation generations
Average
Computed from specs & benchmarks via our public methodology.
Full specifications
Everything we track for the NVIDIA RTX A6000
General
- Announced
- 2020
- Architecture
- Ampere
- TDP
- 300 W
- Power connector
- 1x 8-pin CPU
Cores & Clocks
- Shading units
- 10,752
- RT cores
- 84
- Tensor / AI cores
- 336
- Boost clock
- 1,800 MHz
Memory
- VRAM
- 48 GB
- VRAM type
- GDDR6
- Bus width
- 384 bit
- Bandwidth
- 768 GB/s
Benchmarks
- 3DMark Time Spy
- 17,801Top 25%
- 3DMark Speed Way
- 5,081Top 29%
- PassMark G3D
- 22,606Top 33%
Quick answers
Is the NVIDIA RTX A6000 good in 2026?
The NVIDIA RTX A6000 earns an Inno Score of 55.9/100 (“Average”) based on its specs and benchmark results. The RTX A6000 is NVIDIA's Ampere workstation flagship, with 10,752 CUDA cores and 48 GB of GDDR6 on a 384-bit bus. It targets rendering, simulation and AI work, with NVLink support to pair two cards for 96 GB.
What are the key specs of the NVIDIA RTX A6000?
VRAM: 48 GB · Boost clock: 1,800 MHz · TDP: 300 W · 3DMark Time Spy: 17,801
What are the closest alternatives to the NVIDIA RTX A6000?
Its closest rivals on InnoReviews are NVIDIA GeForce RTX 4070 Super, NVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 4070 Ti — open the comparison pages for a spec-by-spec breakdown.
