GPU comparison

GeForce RTX 5090 vs RTX PRO 6000 Blackwell

Compared on the specs that decide a purchase: memory, bandwidth, power, and which workloads each card actually fits — gaming, creative, or local models when that is the job.

Specifications verified October 2, 2026. Launch MSRP is not today's price — in September 2026 the memory shortage pushed street prices 1.2×–3× above it; the notes row below carries a dated lowest UK/US price where we have one.

NVIDIA

GeForce RTX 5090

32GB VRAM · 1792 GB/s

NVIDIA

RTX PRO 6000 Blackwell

96GB VRAM · 1792 GB/s

Amazon UK search links are provided for current availability; no affiliate programme is currently configured, and no live price is claimed here. Affiliate disclosure.

GPU specification, memory, bandwidth, and price comparison
Specification	GeForce RTX 5090	RTX PRO 6000 Blackwell
Vendor
Vendor	NVIDIA	NVIDIA
Memory
Memory	32GB	Winner: 96GB
Usable for a model
Usable for a model	32GB	Winner: 96GB
Bandwidth
Bandwidth	1792 GB/s	1792 GB/s
Segment
Segment	consumer	workstation
Released
Released	January 30, 2025	Not verifiedUnverified
Launch MSRP
Launch MSRP	~$1,999~£1,481	No launch MSRP published
Notes
Notes	32GB GDDR7; NVIDIA specifies 1792GB/s theoretical memory bandwidth. Launch starting price $1,999 and announced availability January 30, 2025. These figures establish neither current stock nor current purchase price. Local-model fit is a planning estimate; model files, context, cache, backend and driver support determine actual memory use and speed.	Workstation Edition: NVIDIA specifies 96GB GDDR7 ECC, 1792GB/s memory bandwidth and up to 600W. The March 18, 2025 announcement planned distribution from April and OEM systems from May; March 18 is not a verified shipment day. A manufacturer launch MSRP is unverified, so the previous retailer-associated $8,565 value is withheld. Max-Q and Server editions are separate configurations. Memory capacity does not guarantee that every 70B model/runtime/context fits. Verify current purchase or rental terms separately.
UK street price
UK street price	Search Amazon (opens in new tab)	Search Amazon (opens in new tab)

Local models

Estimated local-model fit

Planning estimates from published VRAM and a weights-plus-overhead heuristic. Context length, KV cache, quantization and runtime can change actual memory needs.

Local model fit comparison for both GPUs
Model	GeForce RTX 5090 (est.)	RTX PRO 6000 Blackwell (est.)
Qwen 3 4B4B parameters
Qwen 3 4B4B parameters	Est. fit	Est. fit
Qwen 3.8 27B27B parameters
Qwen 3.8 27B27B parameters	Est. fit	Est. fit
Gemma 4 31B30.7B parameters
Gemma 4 31B30.7B parameters	Est. fit	Est. fit
Llama 3.1 8B Instruct8B parameters
Llama 3.1 8B Instruct8B parameters	Est. fit	Est. fit
Qwen 3 14B14.8B parameters
Qwen 3 14B14.8B parameters	Est. fit	Est. fit
Gemma 3 27B27B parameters
Gemma 3 27B27B parameters	Est. fit	Est. fit
Qwen 3 30B-A3B30.5B parameters
Qwen 3 30B-A3B30.5B parameters	Est. fit	Est. fit
Qwen 3 32B32.8B parameters
Qwen 3 32B32.8B parameters	Est. fit	Est. fit
Llama 3.3 70B70B parameters
Llama 3.3 70B70B parameters	Est. no fit	Est. fit
Muse Glimmer 30B29.6B parameters
Muse Glimmer 30B29.6B parameters	Est. fit	Est. fit
Mixtral 8x22B141B parameters
Mixtral 8x22B141B parameters	Est. no fit	Est. tight
Qwen 3 235B-A22B235B parameters
Qwen 3 235B-A22B235B parameters	Est. no fit	Est. no fit
Qwen 3 Coder 480B-A35B480B parameters
Qwen 3 Coder 480B-A35B480B parameters	Est. no fit	Est. no fit
DeepSeek V4-Flash284B parameters
DeepSeek V4-Flash284B parameters	Est. no fit	Est. no fit

Common questions

GeForce RTX 5090 vs RTX PRO 6000 Blackwell

Answered from the verified figures on this page rather than general guidance.

Which is better for high-refresh gaming, GeForce RTX 5090 or RTX PRO 6000 Blackwell?

Neither has a clear edge on paper — 1792 GB/s against 1792 GB/s is too close to call from specs alone. Compare dated game tests at your target resolution and verify VRAM (32GB usable on GeForce RTX 5090, 96GB on RTX PRO 6000 Blackwell).

Which has more VRAM for gaming and creative work, GeForce RTX 5090 or RTX PRO 6000 Blackwell?

RTX PRO 6000 Blackwell — 96GB usable against 32GB on GeForce RTX 5090. Extra memory helps texture mods, video timelines, and 3D scenes before it helps esports frame rates.

For local inference only — which holds larger models, GeForce RTX 5090 or RTX PRO 6000 Blackwell?

RTX PRO 6000 Blackwell. With 96GB usable against 32GB it additionally holds Llama 3.3 70B, Mixtral 8x22B at Q4.

For local inference only — is GeForce RTX 5090 or RTX PRO 6000 Blackwell faster at generating tokens?

Neither meaningfully — 1792 GB/s against 1792 GB/s is too close to notice. Token throughput will be similar once a model fits, so decide on memory capacity instead.

For local inference only — can either run a 70B model?

RTX PRO 6000 Blackwell can, at Q4 — a 70B model needs about 48GB including runtime overhead. GeForce RTX 5090 does not have the memory. Expect little headroom for long context.

If one card holds a model the other cannot, that capacity difference usually outweighs a bandwidth advantage. A model that spills to system memory can be substantially slower, but the actual impact depends on the backend, transfer path, context, and workload.

All GPU comparisons Compare hosted models instead