GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

Tesla X2070

Graphics processor

Video memory: 6 GB^*
Memory bus: 384 bit^*
Rated memory bandwidth: 177.40 GB/s^*

Benchmark results

LLM Inference: llama.cpp Bench: llama 7B Q4_0 (Q4_0) · version unreported · score14.27^* tokens/sPlatform unreported Compare results
LLM Inference: llama.cpp Bench: llama-2-7b-Q4_0 (Q4_0) · version unreported · score14.27^* tokens/sPlatform unreported Compare results

Processing & platform

Architecture: Fermi^*
Process node: 40 nm^*
GPU cores: 448^*
GPU base clock: 651 MHz^*
Reported GPU frequency: 651 MHz^*
Graphics interface: MXM-B (3.0)^*
Reported chip die area: Reported value 1: 520 mm²^*
Reported value 2: 529 mm²^*
Chip/package TDP: 225 W^*
Transistors: Reported value 1: 3,100 million^*
Reported value 2: 3,000 million^*
Graphics processor: GF100^*
Texture mapping units: 56^*
Raster output units: 48^*
Streaming multiprocessors: 14^*
Chip foundry: TSMC^*
Card or module format: MXM Module^*
Declared GPU TDP: 225 W^*
Suggested system PSU: 550 W^*
External power connectors: None^*
DirectX / feature level: 12 (11_0)^*
OpenGL: 4.6^*
OpenCL: 1.1^*
CUDA compute capability: 2.0^*
Shader model: 5.1^*
GPU L2 cache: 768 KB^*

Memory & storage

Memory options: 6^*
Video memory: 6 GB^*
Video memory type: Reported value 1: GDDR5^*
Reported value 2: GDDR6^*
Reported value 3: GDDR5^*
Reported value 4: GDDR5^*
Reported value 5: GDDR5^*
Reported value 6: GDDR5^*
Memory bus: 384 bit^*
Rated memory bandwidth: 177.40 GB/s^*

Power specifications

Thermal design power: 225 W^*

System configuration

Detailed specifications

Specifications 1 · 12 fields

Foundry: TSMC^*
Vendor: NVIDIA^*
FP32 GFLOPS: 1,332^* GFLOPS
FP64 GFLOPS: 666.1^* GFLOPS
Product: NVIDIA Tesla X2070^*
Type: GPU^*
Release Date: 2011-07-25^*
Process Size (nm): 40^* nm
TDP (W): 225^* W
Die Size (mm^2): 520^* mm²
Transistors (million): 3,000^*
Freq (MHz): 651^* MHz

Specifications 2 · 31 fields

proc transistors

3,100 million^*

proc die size

529^* mm²

cores

448^*

sm count

14^*

memory clock

924^*

ram

6^*

name

NVIDIA Tesla X2070^*

test name Graphics benchmark result

LLM Inference: llama.cpp Bench: llama 7B Q4_0 (Q4_0)^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

test name Graphics benchmark result

LLM Inference: llama.cpp Bench: llama-2-7b-Q4_0 (Q4_0)^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

ram type

GDDR5^*

bus width

384^*

bus interface

MXM-B (3.0)^*

ram bandwidth

177.4^* GB/s

fp32 performance

1.1657^*

fp64 performance

0.5828^*

score Graphics benchmark result

14.27^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

score Graphics benchmark result

14.27^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

tdp

225^* W

suggested psu

550^* W

slot width

MXM Module^*

api cuda

2^*

is retail board

No^*

gpi value

1.77^*

unit Graphics benchmark result

tokens/s^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

unit Graphics benchmark result

tokens/s^*

Scope and conditions

benchmark_context

suite: Per-row third-party test name^*
test_configuration_verified: No^*
comparable: No^*

vendor

nvd^*

manufacturer

NVIDIA^*

arch name

Fermi^*

card release date

2011-07-25^*

proc foundry

TSMC^*

proc process size

40^* nm

Specifications 3 · 36 fields

CUDA compute capability

2.0^*

DirectX / feature level

12 (11_0)^*

OpenCL

1.1^*

OpenGL

4.6^*

Shader model

5.1^*

Architecture

Fermi^*

Graphics interface

MXM-B (3.0)^*

Memory bus

384^* bit

Reported shader processors

448^*

Reported theoretical FP32 peak

1.1657^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.5828^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

64 KB (per SM)^*

Declared GPU L2 cache

768 KB^*

Reported display outputs

No outputs^*

Theoretical pixel fill rate

18.23 GPixel/s^*

External power connectors

None^*

Reported chip die area

529 mm²^*

Chip foundry

TSMC^*

Process node declaration

40 nm^*

Transistors

3,100 million^*

Graphics processor

GF100^*

Reported video memory

6^* GB

Rated memory bandwidth

177.4 GB/s^*

Video memory type

GDDR5^*

Raster output units

48^*

Reported shading units

448^*

Card or module format

MXM Module^*

Streaming multiprocessors

14^*

Suggested system PSU

550 W^*

2:4 tensor sparsity

No^*

Declared GPU TDP

225 W^*

Theoretical texture fill rate

36.46 GTexel/s^*

Texture mapping units

56^*

GPU cores

448^*

Video memory

6^* GB

Video memory type

GDDR5^*

Specifications 4 · 8 fields

Product Name: Tesla X2070^*
GPU Chip: GF100^*
Released: Jul 25th, 2011^*
Bus: MXM-B (3.0)^*
Memory: 6 GB, GDDR5, 384 bit^*
GPU clock: 651^* MHz
Memory clock: 924^* MHz
Shaders / TMUs / ROPs: 448 / 56 / 48^*

Specifications 5 · 51 fields

Brand: NVIDIA^*
Top BUS WIDTH: 384^* bit
Graphics Processor GPU Name: GF100^*
Clock Speeds Shader Clock: 1,301^* MHz
Graphics Processor Architecture: Fermi^*
Render Config SM Count: 14^*
Graphics Features CUDA: 2^*
Graphics Processor Process Size: 40^* nm
Graphics Processor Transistors: 3,100 million^*
Graphics Processor Die Size: 529^* mm²
Graphics Card Release Date: Jul 25th, 2011^*
Graphics Card Generation: Tesla Fermi (x20xx)^*
Graphics Card Successor: Tesla Kepler^*
Graphics Card Production: End-of-life^*
Name: Tesla X2070^*
Graphics Card Bus Interface: MXM-B (3.0)^*
Clock Speeds GPU Clock: 651^* MHz
Clock Speeds Memory Clock: 924 MHz 3.7 Gbps effective^*
Memory Memory Size: 6^* GB
Memory Memory Type: GDDR5^*
Memory Memory Bus: 384^* bit
Memory Bandwidth: 177.4^* GB/s
Render Config TMUs: 56^*
Top GRAPHICS PROCESSOR: GF100^*
Render Config ROPs: 48^*
Theoretical Performance Pixel Rate: 18.23 GPixel/s
Theoretical Performance Texture Rate: 36.46 GTexel/s
Board Design Slot Width: MXM Module^*
Board Design TDP: 225^* W
Board Design Suggested PSU: 550^* W
Board Design Outputs: No outputs^*
Graphics Features DirectX: 12 (11_0)^*
Graphics Features OpenGL: 4.6^*
Graphics Features OpenCL: 1.1^*
source file: nvidia2011.json^*
Graphics Processor Foundry: TSMC^*
Graphics Card Predecessor: Tesla^*
Graphics Processor GPU Variant: GF100-876-A3^*
Graphics Processor Density: 5.9M / mm²^*
Top TMUS: 56^*
Graphics Processor Chip Package: BGA-1980^*
Top ROPS: 48^*
Graphics Features Shader Model: 5.1^*
Top CORES: 448^*
Render Config Shading Units: 448^*
Render Config L2 Cache: 768^* KB
Top MEMORY SIZE: 6^* GB
Theoretical Performance FP32 (float): 1,165.7 GFLOPS
Theoretical Performance FP64 (double): 582.8 GFLOPS (1:2)
Render Config L1 Cache: 64 KB (per SM)^*
Top MEMORY TYPE: GDDR5^*

Specifications 6 · 27 fields

manufacturer: NVIDIA^*
manufacturer: NVIDIA^*
rop: 48^*
rop: 64^*
igp: No^*
igp: No^*
bus: PCIe 2.0 x16^*
bus: PCIe 4.0 x8^*
memType: GDDR5^*
memType: GDDR6^*
gpuChip: Navi 33^*
gpuChip: GF110^*
productName: Tesla X2070^*
productName: Tesla X2070^*
releaseYear: 2,011^*
releaseYear: 2,011^*
memSize: 6^*
memSize: 6^*
memBusWidth: 384^*
gpuClock: 1,720^*
gpuClock: 651^*
memClock: 2,250^*
memClock: 924^*
unifiedShader: 512^*
unifiedShader: 2,048^*
tmu: 64^*
tmu: 128^*

* Provisional data