GeekRadar / The comparison desk
COMPARE.
GPUGeForce RTX 3080 (10 GB)
Graphics processor
- Video memory
- 10 GB*
- Memory bus
- 320 bit*
- Rated memory bandwidth
- 760 GB/s*
No matched benchmark results are available for this entry yet.
Processing & platform
- Architecture
- Samsung 8LPP*
Memory & storage
- Video memory
- 10 GB*
- Video memory type
- GDDR6X*
- Memory bus
- 320 bit*
- Rated memory bandwidth
- 760 GB/s*
System configuration
Detailed specifications
Specifications 1 · 20 fields
- GPU architecture — GeForce RTX 3080 — 10 GB GDDR6X
- Ampere*
- Theoretical peak: BF16 tensor dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 59.5* TFLOPS
Scope and conditions
- Accumulation
- FP32*
- Theoretical peak: FP16 tensor dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 59.5* TFLOPS
Scope and conditions
- Accumulation
- FP32*
- Theoretical peak: FP16 tensor with FP16 accumulation dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 119* TFLOPS
- Theoretical peak: FP32 non-tensor dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 29.8* TFLOPS
- Native hardware support: FP4 tensor — GeForce RTX 3080 — 10 GB GDDR6X
- No*
- Native hardware support: FP6 tensor — GeForce RTX 3080 — 10 GB GDDR6X
- No*
- Native hardware support: FP8 tensor — GeForce RTX 3080 — 10 GB GDDR6X
- No*
- Native hardware support: FP8 tensor with FP16 accumulation — GeForce RTX 3080 — 10 GB GDDR6X
- No*
- Theoretical peak: INT8 tensor dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 238* TOPS
- Theoretical peak: TF32 tensor dense per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 29.8* TFLOPS
- Physical configuration — GeForce RTX 3080 — 10 GB GDDR6X
- Desktop graphics card (PCIe add-in card)*
- Native hardware support for GPU interconnect — GeForce RTX 3080 — 10 GB GDDR6X
- No*
- PCIe interface declaration — GeForce RTX 3080 — 10 GB GDDR6X
- PCIe Gen 4*
- Maximum board power — GeForce RTX 3080 — 10 GB GDDR6X
- 320* W
- Peak memory bandwidth per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 760* GB/s
- Memory interface width per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 320* bit
- Memory capacity per GPU — GeForce RTX 3080 — 10 GB GDDR6X
- 10* GB
- Memory technology — GeForce RTX 3080 — 10 GB GDDR6X
- GDDR6X*
- GPU model — GeForce RTX 3080 — 10 GB GDDR6X
- GeForce RTX 3080*
Specifications 2 · 7 fields
- type
- gpu*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- brand
- NVIDIA*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- model
- GeForce RTX 3080 (10 GB)*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- official power w
- 320* W
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- what the figure is
- Graphics Card Power*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- maker min psu w
- 750*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
- updated
- 2026-09-29*
Scope and conditions
- source_native_row_context
- what_the_figure_is
- Graphics Card Power*
- updated
- 2026-09-29*
- brand
- NVIDIA*
- model
- GeForce RTX 3080 (10 GB)*
- type
- gpu*
Specifications 3 · 25 fields
- model
- GeForce RTX 3080 (10 GB)*
- base clock mhz
- 1,440* MHz
- max boost clock mhz
- 1,710* MHz
- bus type
- GDDR6X*
- bus width
- 320*
- vram
- 10*
- clock speed
- 19,000*
- bandwith
- 760*
- fp16
- 25,060*
- fp32
- 25,070*
- fp64
- 392*
- price
- 399.95*
- tdp
- 320*
- nv link
- No*
- bus interface
- PCIe 4.0 x16*
- L2 Cache (MB)
- 5*
- eol
- Yes*
- eol date
- 9/2023*
- market
- Consumer*
- ecc equipped
- No*
- architecture
- Samsung 8LPP*
- release date
- 9/17/2020*
- tensor cores
- 272*
- tensor core gen
- 3rd Gen (Ampere)*
- cuda cores
- 8,704*
0 selected Compare 0
* Provisional data