VRAM TableGPU fit and local-LLM VRAM for OEM desktops, from published specifications Search by model
Browse categories

NVIDIA RTX 5000 Ada Generation

NVIDIA (specification common to board partners) · 32 GB · catalogue identifier rtx-5000-ada-32gb · last updated 2026-09-25

82 desktops in this database carry a verdict for the NVIDIA RTX 5000 Ada Generation: 1 fits with conditions, 57 do not fit (7 of these are estimates) and 24 are unverified. Fits with conditions: the HP Z2 Tower G9 Workstation. Estimated Conditions most often concern card length (1 desktop), card width (1) and power (1); for this card, its vendor publishes no recommended system power. Of the 57 desktops it does not fit, 32 conflict with a half-height-only slot and 28 with the desktop's maximum card length; the Dell OptiPlex 3060 Small Form Factor (2018) is one of the 32. Estimated Among the 4 cards with 32 GB in this database, its board power of 250 W is within 200–575 W, with 1 lower and 1 higher; its length of 266.7 mm compares with 266.7–304 mm where published. The largest listed model its 32 GB holds at 32k context is Qwen3 32B (Q4_K_M, needs 27.5 GiB).

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 5000 Ada Generation
Board
NVIDIA RTX 5000 Ada Generation (NVIDIA (specification common to board partners))
Card memory
32 GB · A read 2026-09-19 [S1]
Total board power
250 W · A read 2026-09-19 [S1]
Vendor's recommended system power
Not stated by the cited specification
Auxiliary power connector
PCIe CEM5 16-pin · A read 2026-09-19 [S1]
Length
266.7 mm · A read 2026-09-19 [S1]
Height
111.8 mm
Width
Not stated by the cited specification
Slot widths occupied
2 · A read 2026-09-19 [S1]
Low profile
No · A read 2026-09-19 [S1]
CUDA
Yes

Each graded figure above cites: [S1] NVIDIA RTX 5000 Ada Generation specifications (official).

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

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: 1
  • ✕ Does not fit: 57 (7 of these are estimates) Estimated
  • ? Unverified: 24

Conditions shared by the desktops below. Each is stated once here. They come from this card's own records, or are worded the same for more than one desktop. A desktop listed below names the ones it carries by number; the figures particular to that desktop are on its own page.

  1. Card length: this card is 266.7 mm long. Card length: where a desktop publishes no maximum, measure the clearance in front of the slot and compare it with the card. (1 desktop)
  2. Card width: this card occupies 2 slots. Card width: where a desktop publishes no slot-width ceiling, check that there is room beside the x16 slot. (1 desktop)
  3. Power: this card's vendor publishes no recommended system power, so it cannot be set against a desktop's supply ratings. Check the card's own specification sheet. (1 desktop)
  4. Auxiliary power connector: this card needs a 6-pin connector. Auxiliary power connector: where a desktop's manufacturer does not publish what the supply offers, open the machine and look at the cable before buying. (1 desktop)

△ Fits with conditions 1 desktop:

  • confidence A (1): HP Z2 Tower G9 Workstation [1] [2] [3] [4] (The manufacturer offers this card as a factory configuration for this model, so the chassis, the bracket and the supply are established by that shipping fact — for this model alone.)

✕ Does not fit 57 desktops, grouped by the published limit that conflicts with this card's published figures:

? Unverified 24 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.
VerdictMeaning
✓ 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.

  1. 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. (5 rows)
Calculated from the published file size and the model architecture.
ModelOutcome Needs at 32k contextThe arithmetic Confidence
Qwen3 32B Q4_K_M ✓ Fits in VRAM 27.5 GiB 32 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 32 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 32 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 32 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 32 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 32 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 32 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 32 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 32 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 32 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 32 GB of memory is at least the 9.6 GB required at a context length of 32,768 tokens. 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. BD
Qwen3 4B Q4_K_M ✓ Fits in VRAM 7.9 GiB 32 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 32 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 32 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 32 GB of memory is at least the 4.7 GB required at a context length of 32,768 tokens. B

✕ Exceeds VRAM 3 models need more memory at 32k context than this card has, 2 of them published as an estimate. The arithmetic for each is on the model's own page.

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.

Cards whose published memory size is the same, so the local-model outcomes on this page apply to them as well. Whether a given desktop accepts them is a separate question, answered on each card's own page.

Sources

Every document this page draws on, with the date it was read.
DocumentPublisherKindRead on
NVIDIA RTX 5000 Ada Generation specifications (official) NVIDIA Manufacturer official documentation 2026-09-19
NVIDIA Power Guidelines for Workstation Products (DA-07261-001_v15, Jul 2023) NVIDIA primary 2026-09-19

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