Zestawienie procesorów NVIDIA ProViz

Blackwell

Rdzenie

Wydajność

Pamięć

Zasilanie

PCIe
Gen

Uwagi

TFLOPS

RT

Tensor

CUDA

FP64

FP32

GB

GB/s

Typ

Szyna

TGP (W)

Złącze

Adapter (opcja)

B206

RTX PRO 2000

192

4480

21

16

360

GDRR7

128

70

-

-

5.0

VCNRTXPRO2000-PB / SB

B203

RTX PRO 4000

70

280

8960

0.73

47

24

672

GDDR7

192

140

PCIe CEM5 16-pin
(12V-2x6)

1xPCIe 8-pin -> 1xCEM5 16-pin

5.0 x16

VCNRTXPRO4000-PB / SB

34

70

-

-

VCNRTXPRO4000LP-PB / SB

RTX PRO 4500

82

328

10496

0.86

55

32

896

GDDR7

256

200

PCIe CEM5 16-pin
(12V-2x6)

2xPCIe 8-pin -> 1xCEM5 16-pin (930-00030-1633-000)

5.0 x16

VCNRTXPRO4500-PB / SB

B202

RTX PRO 5000

110

440

14080

1.15

74

48

1344

GDDR7

384

300

PCIe CEM5 16-pin
(12V-2x6)

2xPCIe 8-pin -> 1xCEM5 16-pin (930-00030-1633-000)

5.0 x16

VCNRTXPRO5000-PB / SB

RTX PRO 6000 Blackwell
Server Edition

188

752

24064

1.97

117

96

1600

GDDR7

512

400
-600

PCIe CEM5 16-pin
(12V-2x6)

od producenta serwera

5.0 x16

Passive 4.4” H x 10.5” L
FHFL Dual Slot
TCSRTXPRO6000SE-PB

RTX PRO 6000 Blackwell Max-Q Workstation Edition

1.72

110

1792

300

2xPCIe 8-pin -> 1xCEM5 16-pin (930-00030-1633-000)

Active 4.4” H x 10.5” L
FHFL Dual Slot
VCNRTXPRO6000MQ-PB / SB

RTX PRO 6000 Blackwell Workstation Edition

1.97

125

600

4xPCIe 8-pin -> 1xCEM5 16-pin (930-00030-1818-000)

Active (2x) 5.4” H x 12” L
XHFL Dual Slot
VCNRTXPRO6000-PB / SB

ADA

Rdzenie

Wydajność

Pamięć

Zasilanie

PCIe
Gen

Uwagi

TFLOPS

RT

Tensor

CUDA

FP64

FP32

GB

GB/s

Typ

Szyna

TGP (W)

Złącze

Adapter (opcja)

AD107

RTX 2000 Ada

22

88

2816

0.19

12.0

16

224

GDDR6

128

70

-

-

4.0 x8

NVIDIA RTX 2000 Ada

8.9

50

VCNRTX2000EADA-SB

AD104

RTX 4000 Ada

48

192

6144

0.30

19.2

20

320

GDDR6

160

70

-

-

4.0 x16

NVIDIA RTX 4000 Ada SFF

0.42

26,7

360

130

PCIe CEM5 16-pin
(12VHPWR)

1xPCIe 8-pin -> 1xCEM5 16-pin
(930-00030-1656-000)

4.0 x16

NVIDIA RTX 4000 Ada

RTX 4500 Ada

60

240

7680

0.62

39,9

24

432

GDDR6

192

210

2xPCIe 8-pin -> 1xCEM5 16-pin
(930-00030-1633-000)

4.0 x16

NVIDIA RTX 4500 Ada

AD102

RTX 5000 Ada

100

400

12800

1.02

65,3.

32

576

GDDR6

256

250

4.0 x16

NVIDIA RTX 5000 Ada

RTX 5880 Ada

110

440

14080

1.12

69,3

48

960

GDDR6

384

285

4.0 x16

RTX 6000 Ada

142

568

18176

1.49

91.1

300

4.0 x16

NVIDIA RTX 6000 Ada

Ampere

Rdzenie

Wydajność

Pamięć

Zasilanie

PCIe
Gen

Uwagi

TFLOPS

RT

Tensor

CUDA

FP64

FP32

GB

GB/s

Typ

Szyna

TGP (W)

Złącze

Adapter (opcja)

GA107

RTX A400

6

24

768

2.7

4

96

GDDR6

64

50

-

-

4.0 x8

NVIDIA RTX A400

30

VCNRTXA400-PL

RTX A1000

18

72

2304

6.7

8

192

128

50

4.0 x8

NVIDIA RTX A1000

35

VCNRTXA1000-PL

GA106

RTX A2000

26

104

3328

0.12

7.99

12

288

GDDR6

192

70

-

-

4.0 x16

NVIDIA RTX A2000 12GB

6

NVIDIA RTX A2000 6GB

GA104

RTX A4000

48

192

6144

0.71

22.86

16

448

GDDR6

256

140

PCIe 6-pin

-

4.0 x16

NVIDIA RTX A4000

GA103

RTX A5000

64

256

8192

0.87

27.80

24

768

GDDR6

384

230

PCIe 8-pin

-

4.0 x16

NVIDIA RTX A5000

GA102

RTX A4500

56

224

7168

0.74

23.65

20

640

GDDR6

320

200

PCIe 8-pin

-

4.0 x16

NVIDIA RTX A4500

RTX A5500

80

320

10240

1.09

34.71

24

768

GDDR6

384

230

PCIe 8-pin

-

4.0 x16

NVIDIA RTX A5500

RTX A6000

84

336

10752

1.25

38.71

48

768

GDDR6

384

300

EPS 8-pin

2xPCIe 8-pin -> 1xEPS 8-pin
(930-00030-1233-000)

4.0 x16

NVIDIA RTX A6000

GA100

RTX A800

-

432

6912

9.7

19.5

40

1555

HBM2

5120

240

PCIe CEM5 16-pin

4.0 x16

NVIDIA RTX A800

(brak wyjść wideo)

Turing

Rdzenie

Wydajność

Pamięć

Zasilanie

PCIe
Gen

Uwagi

TFLOPS

RT

Tensor

CUDA

FP64

FP32

GB

GB/s

Typ

Szyna

TGP (W)

Złącze

Adapter (opcja)

TU117

T400

-

-

384

-

1.1

4

80

GDDR6

64

30

-

-

3.0 x16

NVIDIA T400 4GB

2

3.0 x16

NVIDIA T400

T600

-

-

640

-

1.8

4

160

GDDR6

128

40

-

-

3.0 x16

NVIDIA T600

T1000

-

-

896

-

2.5

8

160

GDDR6

128

50

-

-

3.0 x16

NVIDIA T1000 8GB

4

3.0 x16

NVIDIA T1000

TU106

Quadro RTX 4000

36

288

2304

-

7.1

8

416

GDDR6

256

160

PCIe 6-pin

3.0 x16

PNY Quadro RTX 4000

TU104

Quadro RTX 5000

48

384

3072

-

11.2

16

448

265

PCIe 8-pin

3.0 x16

PNY Quadro RTX 5000

TU102

Quadro RTX 8000

72

576

4608

-

16.3

48

670

GDDR6

384

295

PCIe 8-pin + 6-pin

3.0 x16

PNY Quadro RTX 8000

-

14.9

624

250

EPS 8-pin

3.0 x16

PNY Quadro RTX 8000 Passive

Quadro RTX 6000

-

16.3

24

670

295

PCIe 8-pin + 6-pin

3.0 x16

PNY Quadro RTX 6000

-

14.9

624

250

EPS 8-pin

3.0 x16

PNY Quadro RTX 6000 Passive

GV100

Quadro GV100

-

640

5120

7.4

14.8

32

868

HBM2

4096

250

EPS 8-pin

3.0 x16

PNY Quadro GV100