LCM DreamShaper v7 · 8 GPUs measured first-party · text-to-image · Updated October 2026
LCM DreamShaper v7 on 8 GPUs, measured first-party: NVIDIA GeForce RTX 4090 leads at 260.9 images/min, RTX 3060 trails at 62.2 images/min, and it peaked at 5GB of VRAM.
Benchmarked weights: SimianLuo/LCM_Dreamshaper_v7

260.9 images/min on LCM DreamShaper v7, the ceiling. Measured on our bench. 24GB of VRAM, $1,599 at launch.

179.8 images/min on LCM DreamShaper v7, fastest card you can buy at retail. Measured on our bench. 16GB of VRAM, $749 at launch.

62.2 images/min on LCM DreamShaper v7, lowest launch price that still fits. Measured on our bench. 12GB of VRAM, $329 at launch.

142.4 images/min on LCM DreamShaper v7, most speed per dollar. Measured on our bench. 12GB of VRAM, $549 at launch. That is 259.3 images/min per $1,000 of launch price.
What GPU Do You Need for LCM DreamShaper v7?, images/min by GPU
Efficiency: images/min per 100W drawn
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
Launch price, not street price, so it ages. A speed leaderboard always crowns the most expensive card; this is the counterweight.
LCM DreamShaper v7. Measured image generation speed by GPU
| GPU | Images/min | s per image | img/W·min | Avg power |
|---|---|---|---|---|
| NVIDIA GeForce RTX 4090 | 260.9 | 0.23 | 1.47 | 177.5 W |
| GeForce RTX 5070 Ti | 179.8 | 0.334 | 1.157 | 155.4 W |
| GeForce RTX 5070 | 142.4 | 0.421 | 0.988 | 144.1 W |
| GeForce RTX 5060 Ti | 106.2 | 0.565 | 0.923 | 115.0 W |
| NVIDIA GeForce RTX 4060 Ti 16GB | 94.29 | 0.636 | 0.837 | 112.6 W |
| NVIDIA GeForce RTX 3070 Founders Edition | 92.56 | 0.648 | 0.57 | 162.4 W |
| NVIDIA GeForce RTX 3060 Ti | 80.24 | 0.748 | 0.501 | 160.2 W |
| NVIDIA GeForce RTX 3060 | 62.2 | 0.965 | 0.469 | 132.5 W |
What the numbers show. Across 8 GPUs measured on our own bench, RTX 4090 is fastest at 261 images/min. The slowest, RTX 3060, manages 62.2, so the spread is 4.2x from top to bottom. Per dollar of launch price, RTX 5060 Ti gives the most (322.8 images/min per $1,000). The fastest card with 16GB or less is RTX 5070 Ti at 180 images/min.
About LCM DreamShaper v7. LCM DreamShaper v7: from SimianLuo, 0.9B parameters, on Hugging Face since October 2023, MIT licence. 151,494 downloads in the last 30 days.
How it compares. RTX 4090: LCM DreamShaper v7 260.9 images/min, Stable Diffusion 1.5 80.14, SD Turbo 786.1, Stable Diffusion 2.1 60.58, PixArt-Sigma XL 24.27. 1 of 4 beat LCM DreamShaper v7 here.
Cost on a rented GPU. 1,000 images of LCM DreamShaper v7: $0.010 on a RTX 3060 ($0.036/hr, 16 min), $0.021 on a RTX 4090 ($0.34/hr, 4 min, 2.2x the cost).
LCM DreamShaper v7: cost per 1,000 images on rented GPUs
| GPU | Cheapest rate | Speed (images/min) | Cost per 1,000 images |
|---|---|---|---|
| NVIDIA GeForce RTX 3060 | $0.036/hr | 62.2 | $0.010 |
| NVIDIA GeForce RTX 3070 Founders Edition | $0.075/hr | 92.56 | $0.014 |
| GeForce RTX 5070 Ti | $0.15/hr | 179.8 | $0.014 |
| GeForce RTX 5060 Ti | $0.14/hr | 106.2 | $0.021 |
| NVIDIA GeForce RTX 4090 | $0.34/hr | 260.9 | $0.021 |
| GeForce RTX 5070 | $0.19/hr | 142.4 | $0.022 |
Cheapest hourly rate we track on RunPod and Vast.ai, divided by the measured speed. Startup time and storage are extra.
Speed tiers for LCM DreamShaper v7. 30+ images/min: 8 (RTX 4090, RTX 5070 Ti, RTX 5070). 30 images/min means two seconds or less per picture.
Time per image. LCM DreamShaper v7 at 768px and 4 steps: 0.2s per image on the RTX 4090, 1.0s on the RTX 3060. A batch of 100 takes 0 min on the fastest card and 2 min on the slowest.
VRAM for LCM DreamShaper v7. Measured peak 4.7GB, so 8GB is the smallest common card size; smallest card it ran on: RTX 3070 Founders Edition (8GB).
Power on LCM DreamShaper v7. Most efficient: RTX 4090, 178W, 11.3 Wh per 1,000 images.
Fastest on LCM DreamShaper v7: NVIDIA GeForce RTX 4090, 260.9 images/min. Cheapest consumer card that ran it: NVIDIA GeForce RTX 3060 ($329, 62.2 images/min). Cheapest to rent per job: NVIDIA GeForce RTX 3060, $0.010 per 1,000 images.
LCM DreamShaper v7 at 768px 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.