NVIDIA RTX PRO 6000 Blackwell Workstation Edition

NVIDIA (specification common to board partners) · 96 GB · catalogue identifier rtx-pro-6000-blackwell-96gb · last updated 2026-09-25

82 desktops in this database carry a verdict for the NVIDIA RTX PRO 6000 Blackwell Workstation Edition: 75 do not fit (7 of these are estimates) and 7 are unverified. Estimated Of the 75 desktops it does not fit, 74 conflict with the desktop's largest supply option and 33 with the desktop's maximum card length; the Dell OptiPlex 3050 Small Form Factor (2017) is one of the 74. Estimated No card listed here has more memory; the nearest size is 80 GB. The largest listed model its 96 GB holds at 32k context is GLM-4.5-Air (Q4_K_M, needs 74.8 GiB). Estimated

Published specifications for this card, the OEM desktops whose own specifications accept it, and how much of a local language model its memory holds.

We have not physically tested this combination. Data is compiled from official specification sheets and cited sources.

Specifications

Graphics processor: NVIDIA RTX PRO 6000 Blackwell Workstation Edition
Board: NVIDIA RTX PRO 6000 Blackwell Workstation Edition (NVIDIA (specification common to board partners))
Card memory: 96 GB · A read 2026-08-20 [S1]
Total board power: 600 W · A read 2026-08-20 [S2]
Vendor's recommended system power: Not stated by the cited specification
Auxiliary power connector: PCIe CEM5 16-pin · A read 2026-08-20 [S2]
Length: 304.8 mm · A read 2026-08-20 [S1]
Height: Not stated by the cited specification
Width: 137.16 mm
Slot widths occupied: 2 · A read 2026-08-20 [S1]
Low profile: No · A read 2026-08-20 [S1]
CUDA: Yes

Each graded figure above cites: [S1] NVIDIA RTX PRO 6000 Blackwell Workstation Edition specifications (official); [S2] NVIDIA RTX PRO 6000 Blackwell Workstation Edition Datasheet.

Power. Leave headroom; OEM PSUs are often rated at the limit. The recommended system figure above is the card vendor's own, and it assumes a retail supply rather than an OEM one.

Desktops that take this card

No desktop in this database carries a fitting verdict for this card. It was assessed against 82 of them. Its published board power is 600 W. The highest board power among cards that do carry a fitting verdict on a desktop here is 300 W. Its published length is 304.8 mm. The longest card any desktop here publishes a clearance for is 268 mm.

This card has been assessed against 82 desktops in this database. Each verdict is a statement about published specifications, not about a tested combination.

✓ Fits: 0
△ Fits with conditions: 0
✕ Does not fit: 75 (7 of these are estimates) Estimated
? Unverified: 7

✕ Does not fit 75 desktops, grouped by the published limit that conflicts with this card's published figures:

Card length (33). Published maximum card length on these desktops: 147–268 mm. This card is 304.8 mm long. Some also conflict on bracket height (19), power-supply wattage (32), auxiliary power connector (3).
- confidence B (33): HP Elite Tower 800 G9; Lenovo ThinkCentre M70s, ThinkCentre M70s Gen 3, ThinkCentre M70s Gen 4, ThinkCentre M70s Gen 5, ThinkCentre M70t, ThinkCentre M70t Gen 4, ThinkCentre M70t Gen 5, ThinkCentre M710 Small Form Factor, ThinkCentre M710 Tower, ThinkCentre M710e SFF, ThinkCentre M720 SFF, ThinkCentre M720 Tower, ThinkCentre M75s, ThinkCentre M75s Gen 5, ThinkCentre M75t Gen 5, ThinkCentre M80s, ThinkCentre M80s Gen 3, ThinkCentre M80t, ThinkCentre M80t Gen 3, ThinkCentre M90s, ThinkCentre M90s Gen 3, ThinkCentre M90s Gen 4, ThinkCentre M90s Gen 5, ThinkCentre M90t, ThinkCentre M90t Gen 3, ThinkCentre M90t Gen 4, ThinkCentre M90t Gen 5, ThinkCentre M910 Small Form Factor, ThinkCentre M910 Tower, ThinkCentre M920 Small Form Factor, ThinkCentre M920 Tower, ThinkCentre neo 50s Gen 4
Slot width (2, 2 published as an estimate). Published slot-width ceiling on these desktops: 1 slot. This card occupies 2 slots. Estimated Some also conflict on bracket height (2), power-supply wattage (2).
- Estimated confidence BD (1): HP Z2 Small Form Factor G5 Workstation
- Estimated confidence BDE (1): HP Z2 G9 SFF Workstation Desktop PC
Bracket height (half-height slot) (11, 1 published as an estimate). These desktops are recorded as taking only half-height cards. This card is a full-height card. Estimated Some also conflict on power-supply wattage (11).
- confidence B (10): Dell OptiPlex 3060 Small Form Factor (2018), OptiPlex 3090 Small Form Factor (2021), OptiPlex 5040 Small Form Factor (2015), OptiPlex 5080 Small Form Factor (2020), OptiPlex 7000 Small Form Factor (2022), OptiPlex 7040 Small Form Factor (2015), OptiPlex 7080 Small Form Factor (2020), OptiPlex 7090 Small Form Factor (2021), OptiPlex Small Form Factor 7010 (2023), OptiPlex Small Form Factor Plus 7010 (2023)
- Estimated confidence BD (1): HP Z2 Small Form Factor G8 Workstation
Power-supply wattage (29, 4 published as an estimate). Largest published supply option on these desktops: 180–500 W. This card's board power is 600 W. Estimated Some also conflict on auxiliary power connector (7).
- confidence B (25): Dell OptiPlex 3050 Small Form Factor (2017), OptiPlex 3050 Tower (2017), OptiPlex 3060 Tower (2018), OptiPlex 3080 Tower (2020), OptiPlex 3090 Tower (2021), OptiPlex 5040 Mini Tower (2015), OptiPlex 5050 Small Form Factor (2017), OptiPlex 5050 Tower (2017), OptiPlex 5060 Small Form Factor (2018), OptiPlex 5060 Tower (2018), OptiPlex 5080 Tower (2020), OptiPlex 5090 Tower (2021), OptiPlex 7000 Tower (2022), OptiPlex 7040 Mini Tower (2015), OptiPlex 7050 Small Form Factor (2017), OptiPlex 7050 Tower (2017), OptiPlex 7060 Small Form Factor (2018), OptiPlex 7070 Small Form Factor (2019), OptiPlex 7080 Tower (2020), OptiPlex 7090 Tower (2021), OptiPlex Tower 7010 (2023), OptiPlex Tower Plus 7010 (2023); HP Elite Tower 600 G9, Pro Tower 400 G9; Lenovo ThinkCentre neo 50t Gen 4
- Estimated confidence BD (4): Dell OptiPlex 5070 Small Form Factor (2019), OptiPlex 5070 Tower (2019), OptiPlex 7060 Tower (2018), OptiPlex 7070 Tower (2019)

? Unverified 7 desktops. The evidence does not settle these either way, which is not the same as them being ruled out.

Verify with the manufacturer before purchasing.

What each verdict means. Every one of them is a statement about published specifications.
Verdict	Meaning
✓ Fits	Published specifications show no physical or electrical conflict.
△ Fits with conditions	Published specifications allow it only under the stated conditions.
✕ Does not fit	Published specifications rule the combination out: the card needs a connector the machine does not have, draws more than the supply provides, or is physically too large. It is a statement about specifications, not a report of a failed attempt.
? Unverified	There is not enough evidence to say.

A dashed Estimated badge at the start of a line marks every verdict on that line as published as an estimate, and the grade beside it applies to each of them. The document behind each verdict, and the desktop's own figures, are listed on the desktop's own page. This card has a verdict for 82 of the 82 desktops in this database; the remaining 0 have not been assessed against it, which is not the same as them being ruled out.

Local language models this card's memory holds

Arithmetic shared by the models below. Sentences that read the same for more than one model are stated once here; each row names the ones it carries by number.

This outcome is no stronger than the figures it was calculated from: the size of the model file that is distributed is recorded here from third-party or community material, so grade D is shown alongside B. (6 rows)
This outcome is no stronger than the figures it was calculated from: the size of the model file that is distributed and this model's layer count and this model's key/value head count and this model's head dimension are recorded here from third-party or community material, so grade D is shown alongside B. (2 rows)

Calculated from the published file size and the model architecture.
Model	Outcome	Needs at 32k context	The arithmetic	Confidence
GLM-4.5-Air Q4_K_M	✓ Fits in VRAM Estimated	74.8 GiB	96 GB of memory is at least the 74.8 GB required at a context length of 32,768 tokens. [1]	BD
GPT-OSS 120B MXFP4 (the vendor's own native quantisation)	✓ Fits in VRAM	62.9 GiB	96 GB of memory is at least the 62.9 GB required at a context length of 32,768 tokens.	B
Llama 3.1 70B Q4_K_M	✓ Fits in VRAM Estimated	50.7 GiB	96 GB of memory is at least the 50.7 GB required at a context length of 32,768 tokens. [2]	BD
Qwen3 32B Q4_K_M	✓ Fits in VRAM	27.5 GiB	96 GB of memory is at least the 27.5 GB required at a context length of 32,768 tokens.	B
EXAONE 4.0 32B Q4_K_M	✓ Fits in VRAM	21.8 GiB	96 GB of memory is at least the 21.8 GB required at a context length of 32,768 tokens.	B
Qwen3 30B-A3B Instruct 2507 Q4_K_M	✓ Fits in VRAM Estimated	21.3 GiB	96 GB of memory is at least the 21.3 GB required at a context length of 32,768 tokens. [1]	BD
Gemma 3 27B Q4_0 (QAT)	✓ Fits in VRAM	20.0 GiB	96 GB of memory is at least the 20.0 GB required at a context length of 32,768 tokens.	B
Mistral Small 3.2 24B Instruct Q4_K_M	✓ Fits in VRAM Estimated	19.4 GiB	96 GB of memory is at least the 19.4 GB required at a context length of 32,768 tokens. [1]	BD
Phi-4 14B Q4_K_M	✓ Fits in VRAM Estimated	15.6 GiB	96 GB of memory is at least the 15.6 GB required at a context length of 32,768 tokens. [1]	BD
GPT-OSS 20B MXFP4 (the vendor's own native quantisation)	✓ Fits in VRAM	14.6 GiB	96 GB of memory is at least the 14.6 GB required at a context length of 32,768 tokens.	B
Qwen3 14B Q4_K_M	✓ Fits in VRAM	14.4 GiB	96 GB of memory is at least the 14.4 GB required at a context length of 32,768 tokens.	B
Gemma 3 12B Q4_0 (QAT)	✓ Fits in VRAM	10.9 GiB	96 GB of memory is at least the 10.9 GB required at a context length of 32,768 tokens.	B
Qwen3 8B Q4_K_M	✓ Fits in VRAM	10.2 GiB	96 GB of memory is at least the 10.2 GB required at a context length of 32,768 tokens.	B
Llama 3.1 8B Q4_K_M	✓ Fits in VRAM Estimated	9.6 GiB	96 GB of memory is at least the 9.6 GB required at a context length of 32,768 tokens. [2]	BD
Qwen3 4B Q4_K_M	✓ Fits in VRAM	7.9 GiB	96 GB of memory is at least the 7.9 GB required at a context length of 32,768 tokens.	B
Qwen3 1.7B Q4_K_M	✓ Fits in VRAM Estimated	5.6 GiB	96 GB of memory is at least the 5.6 GB required at a context length of 32,768 tokens. [1]	BD
Qwen3 0.6B Q4_K_M	✓ Fits in VRAM Estimated	4.9 GiB	96 GB of memory is at least the 4.9 GB required at a context length of 32,768 tokens. [1]	BD
Gemma 3 4B Q4_0 (QAT)	✓ Fits in VRAM	4.7 GiB	96 GB of memory is at least the 4.7 GB required at a context length of 32,768 tokens.	B

Exceeds VRAM does not mean the model will not run: inference software can keep part of a model in system memory. How fast that is, and whether it is pleasant to use, is outside what this site measures. The sizing rule and its constants are on the Methodology page.

Sources

Every document this page draws on, with the date it was read.
Document	Publisher	Kind	Read on
NVIDIA RTX PRO 6000 Blackwell Workstation Edition specifications (official)	NVIDIA	Manufacturer official documentation	2026-08-20
NVIDIA RTX PRO 6000 Blackwell Workstation Edition Datasheet	NVIDIA	primary	2026-08-20

We have not physically tested this combination. Data is compiled from official specification sheets and cited sources. Verify with the manufacturer before purchasing.