Qwen-Image-Edit 2509 · 4 GPUs measured first-party · text-to-image · Updated October 2026

What GPU Do You Need for Qwen-Image-Edit 2509?

Qwen-Image-Edit 2509 on 4 GPUs, measured first-party: NVIDIA H200 leads at 1.71 images/min, A100 80GB SXM4 trails at 0.75 images/min, and it peaked at 63GB of VRAM.

Benchmarked weights: Qwen/Qwen-Image-Edit-2509

Fastest we measured
NVIDIA H200

NVIDIA H200

1.71 images/min on Qwen-Image-Edit 2509, the ceiling. Measured on our bench. 141GB of VRAM, $31,000 at launch.

Pros
  • 1.71 images/min on Qwen-Image-Edit 2509
  • 141GB, clears the Qwen-Image-Edit 2509 floor
  • Rentable by the hour rather than bought
Cons
  • 700W board rating
  • Datacenter or workstation hardware, not a retail purchase
Cheapest card that runs it
NVIDIA RTX PRO 6000 Blackwell Workstation Edition

NVIDIA RTX PRO 6000 Blackwell Workstation Edition

1.13 images/min on Qwen-Image-Edit 2509, lowest launch price that still fits. Measured on our bench. 96GB of VRAM, $8,565 at launch.

Pros
  • 1.13 images/min on Qwen-Image-Edit 2509
  • 96GB, clears the Qwen-Image-Edit 2509 floor
  • Rentable by the hour rather than bought
Cons
  • 600W board rating
  • Datacenter or workstation hardware, not a retail purchase
1.71images/min
Fastest: NVIDIA H200
measured, 3-run average
~63GB
VRAM needed (measured peak)
GPU-independent, applies to every card
4
GPUs measured
same pinned harness
0.002images/min/W
Most efficient: NVIDIA H200
real power sampling, not TDP

What GPU Do You Need for Qwen-Image-Edit 2509?, images/min by GPU

NVIDIA H200
1.71 images/min
NVIDIA H100 80GB HBM3
1.65 images/min
NVIDIA RTX PRO 6000 Blackwell Workstation Edition
1.13 images/min
NVIDIA A100 80GB SXM4
0.75 images/min

Efficiency: images/min per 100W drawn

NVIDIA H200
0.25 images/min / 100W
NVIDIA H100 80GB HBM3
0.24 images/min / 100W
NVIDIA RTX PRO 6000 Blackwell Workstation Edition
0.2 images/min / 100W
NVIDIA A100 80GB SXM4
0.19 images/min / 100W

Power is the average pulled during the run, sampled at 1Hz. The fastest card is often not the one here, and for anything left running this is the number that shows up on the bill.

Value: images/min per $1,000 of MSRP

NVIDIA RTX PRO 6000 Blackwell Workstation Edition
0.13 images/min / $1k
NVIDIA H200
0.06 images/min / $1k
NVIDIA H100 80GB HBM3
0.05 images/min / $1k
NVIDIA A100 80GB SXM4
0.04 images/min / $1k

Launch price, not street price, so it ages. A speed leaderboard always crowns the most expensive card; this is the counterweight.

Qwen-Image-Edit 2509. Measured image generation speed by GPU

NVIDIA H2001.71
NVIDIA H100 80GB HBM31.65
NVIDIA RTX PRO 6000 Blackwell Workstation Edition1.13
NVIDIA A100 80GB SXM40.75
GPUImages/mins per imageimg/W·minAvg power
NVIDIA H2001.7135.190.002689.8 W
NVIDIA H100 80GB HBM31.6536.310.002688.5 W
NVIDIA RTX PRO 6000 Blackwell Workstation Edition1.1353.10.002569.8 W
NVIDIA A100 80GB SXM40.7579.50.002395.8 W

What the numbers show. Across 4 GPUs measured on our own bench, H200 is fastest at 1.71 images/min. The slowest, A100 80GB SXM4, manages 0.75, so the spread is 2.3x from top to bottom. Per dollar of launch price, RTX PRO 6000 Blackwell Workstation Edition gives the most (0.1 images/min per $1,000).

How it compares. H100 80GB HBM3: Qwen-Image-Edit 2509 1.65 images/min, Qwen-Image 3.0 (20B), Qwen-Image 2512 3.0 (20B), Z-Image 3.58 (6B), Qwen-Image-Edit 3.68 (20B). All 4 beat Qwen-Image-Edit 2509 here.

Cost on a rented GPU. 1,000 images of Qwen-Image-Edit 2509: $15.87 on a RTX PRO 6000 Blackwell Workstation Edition ($1.08/hr, 14.7 hours), $34.99 on a H200 ($3.59/hr, 9.7 hours, 2.2x the cost).

Qwen-Image-Edit 2509: cost per 1,000 images on rented GPUs

NVIDIA RTX PRO 6000 Blackwell Workstation Edition$1.08/hr
NVIDIA A100 80GB SXM4$0.95/hr
NVIDIA H100 80GB HBM3$2.14/hr
NVIDIA H200$3.59/hr
GPUCheapest rateSpeed (images/min)Cost per 1,000 images
NVIDIA RTX PRO 6000 Blackwell Workstation Edition$1.08/hr1.13$15.87
NVIDIA A100 80GB SXM4$0.95/hr0.75$21.04
NVIDIA H100 80GB HBM3$2.14/hr1.65$21.58
NVIDIA H200$3.59/hr1.71$34.99

Cheapest hourly rate we track on RunPod and Vast.ai, divided by the measured speed. Startup time and storage are extra.

Speed tiers for Qwen-Image-Edit 2509. under 6 images/min: 4 (H200, H100 80GB HBM3, RTX PRO 6000 Blackwell Workstation Edition). 30 images/min means two seconds or less per picture.

Time per image. Qwen-Image-Edit 2509 at 1024px and 40 steps: 35.1s per image on the H200, 80.0s on the A100 80GB SXM4. A batch of 100 takes 58 min on the fastest card and 133 min on the slowest.

VRAM for Qwen-Image-Edit 2509. Measured peak 59.4GB, so 80GB is the smallest common card size; smallest card it ran on: H100 80GB HBM3 (80GB).

Power on Qwen-Image-Edit 2509. Most efficient: H200, 690W, 6.72 kWh per 1,000 images. At $0.15/kWh: $1.01 per 1,000 images.

Our verdict

Fastest on Qwen-Image-Edit 2509: NVIDIA H200, 1.71 images/min. Cheapest to rent per job: NVIDIA RTX PRO 6000 Blackwell Workstation Edition, $15.87 per 1,000 images.

FAQ

What GPU do I need to run Qwen-Image-Edit 2509?
About 62GB. Smallest card that ran it: NVIDIA H100 80GB HBM3 (80GB).
How much does it cost to run Qwen-Image-Edit 2509 in the cloud?
$15.87 per 1,000 images on a NVIDIA RTX PRO 6000 Blackwell Workstation Edition at $1.08/hr, cheapest of 4 rentable cards we measured.
Can I run Qwen-Image-Edit 2509 on a 12GB, 16GB or 24GB card?
It used 59.4GB at the precision we tested. 12GB: no; 16GB: no; 24GB: no.
Is the H100 80GB HBM3 or the A100 80GB SXM4 faster for Qwen-Image-Edit 2509?
The H100 80GB HBM3: 1.65 vs 0.75 images/min, 120% faster on our bench.

How we test

Qwen-Image-Edit 2509 at 1024px in diffusers, bf16, native precision with no offload, timed over three generations after a warmup, with power and VRAM sampled throughout. Image speed tracks tensor compute and architecture generation more than memory bandwidth, so the order here differs from our LLM boards.