Pops' Stable Diffusion Speed List
A hand curated list of generation speeds for various hardware and models.
LAST UPDATED: 2025.08.09
Use the ComfyUI workflow above to start testing.
Contributor List
- Mr Pops Alot (your's truely): RTX 3090, GTX 1080 Ti, GTX 980, GTX 1050 Ti, Pro W5500, Pro Wx 4100, HD 7790, HD 7750, MX 330, 5800X, 10700K, 1165G7, Q9300, T9300, 4790K, 6300U, 3770, 4300U, 5257U
- Lopi99: PRO 6000 B, RTX 5060 Ti 16GB, A4000, 4060 Ti 16GB, 3060 Ti, EPYC 9654
- Hugs288: RTX 4090
- Disty: RX 7900 XTX, A770, 5800X3D, Dimensity 1080
- Panchovix: 5090 (Fedora 42)
Methodology
All GPUs tested use the following settings:
- Euler sampler
- Normal scheduler
- CFG 8
- Square aspect ratio (1:1)
- Steps can be changed. 20 is fine for slow systems but for systems with high end modern GPUs you may have to increase the step count to 100 or more for accurate readings.
- OS if known will be stated. The OS used for initial testing is Arch Linux (Unstated OSes in submissions will be considered as "Unknown"). DirectML stuff is all done on Windows 11 22H2
- Attention is default (unset), unless stated otherwise.
- The workflow used is the provided workflow in the repo.
- App versions are listed. If no version is listed then the commit version will be "Unknown".
- SD1.5: jzli/Hassaku-1.3
- SDXL: OnomaAIResearch/Illustrious-XL-v2.0
- Lumina 2: neta-art/NetaLumina_Alpha Round NNNN EP6 S127716
Raw gen times are not recorded due to variance due to steps being variable. Instead iterations per second (and the inverse of it) are given since they are independent of steps.
The given speed value (it/s or s/it) is used, and then extrapolated using the formula 1/speed to get the other value. If its under 0.01 then it will be expanded to four digits compared to the usual 2
If you can contribute to the list, do so as well. Lets make the most comprehensive, curated list of local Image Gen speeds!
Models Used
The following are the models used for testing. The models you use can be the same architecture as the tested models
Benchmarks
Lumina 2
1536px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 1.88it/s | 0.53s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 1.31it/s | 0.76s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 4090 | 1.29it/s | 0.78s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 3090 | 0.41it/s | 2.40s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| AMD RX 7900 XTX | 0.34it/s | 2.98s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| AMD RX 7900 XTX | 0.33it/s | 3.06s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX A4000 | 0.32it/s | 3.11s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5060 Ti | 0.26it/s | 3.83s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX 3070 | 0.25it/s | 3.95s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | CPU TE |
| NVIDIA RTX 3060 Ti | 0.22it/s | 4.57s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | CPU TE |
| NVIDIA RTX 3060 | 0.18it/s | 5.47s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 | 0.18it/s | 5.52s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | CPU TE |
| Intel A770 | 0.11it/s | 9.38s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 1080 Ti | 0.0421it/s | 23.77s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux |
1024px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 5.12it/s | 0.20s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 3.70it/s | 0.27s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 5090 | 2.58it/s | 0.39s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 4090 | 2.22it/s | 0.45s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 3090 | 1.14it/s | 0.87s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | Sage Attention |
| AMD RX 7900 XTX | 1.08it/s | 0.96s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| AMD RX 7900 XTX | 1.04it/s | 0.96s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3090 | 1.00it/s | 1.00s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 0.82it/s | 1.22s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5060 Ti | 0.68it/s | 1.47s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX 3070 | 0.66it/s | 1.52s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 Ti | 0.58it/s | 1.72s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3060 | 0.44it/s | 2.25s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 | 0.44it/s | 2.28s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | CPU TE |
| Intel A770 | 0.24it/s | 4.24s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 1080 Ti | 0.15it/s | 6.77s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA GTX 980 | 0.0599it/s | 16.69s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | FP32 CPU TE |
| AMD Ryzen 5800X | 0.0102it/s | 97.86s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.0090it/s | 107.25s/it | CPU | ComfyUI (Unknown) | Arch Linux |
512px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 17.43it/s | 0.0574s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 12.17it/s | 0.0821s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 5090 | 8.85it/s | 0.11s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 5090 | 8.04it/s | 0.12s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| AMD RX 7900 XTX | 4.65it/s | 0.22s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| NVIDIA RTX 3090 | 4.35it/s | 0.23s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | Sage Attention |
| AMD RX 7900 XTX | 4.18it/s | 0.24s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX A4000 | 3.29it/s | 0.30s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 5060 Ti | 3.22it/s | 0.31s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX 3070 | 2.62it/s | 0.38s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 Ti | 2.28it/s | 0.44s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3060 | 1.65it/s | 0.61s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | CPU TE |
| NVIDIA RTX 3060 | 1.44it/s | 0.69s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA GTX 1080 Ti | 0.73it/s | 1.37s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| Intel A770 | 0.58it/s | 1.75s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 980 | 0.28it/s | 3.57s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | FP8 CPU TE |
| NVIDIA GTX 980 | 0.25it/s | 3.99s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | FP32 CPU TE |
| AMD Ryzen 5800X | 0.0649it/s | 15.42s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.0513it/s | 19.49s/it | CPU | ComfyUI (Unknown) | Arch Linux |
256px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 18.78it/s | 0.0532s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 18.29it/s | 0.0547s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 4090 | 14.92it/s | 0.067s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 5090 | 13.42it/s | 0.0745s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 3090 | 11.37it/s | 0.0880s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | Sage Attention |
| NVIDIA RTX A4000 | 10.22it/s | 0.10s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| AMD RX 7900 XTX | 9.71it/s | 0.10s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3070 | 9.14it/s | 0.11s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 5060 Ti | 8.76it/s | 0.11s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| AMD RX 7900 XTX | 7.93it/s | 0.13s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| NVIDIA RTX 3060 Ti | 7.21it/s | 0.14s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA GTX 1080 Ti | 1.74it/s | 0.57s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA GTX 980 | 0.78it/s | 1.27s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | FP8 CPU TE |
| Intel A770 | 0.71it/s | 1.40s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 980 | 0.59it/s | 1.68s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | FP32 CPU TE |
| Intel i7-10700K | 0.26it/s | 3.78s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| AMD Ryzen 5800X | 0.25it/s | 3.98s/it | CPU | ComfyUI (Unknown) | Arch Linux |
SDXL
1536px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 6.61it/s | 0.15s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 6.12it/s | 0.16s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 5090 | 3.38it/s | 0.29s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 4090 | 3.11it/s | 0.32s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| AMD RX 7900 XTX | 1.96it/s | 0.51s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| AMD RX 7900 XTX | 1.93it/s | 0.52s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3090 | 1.63it/s | 0.61s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 1.16it/s | 0.86s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| Intel A770 | 1.12it/s | 0.89s/it | OpenVINO 2025.2 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 5060 Ti | 1.10it/s | 0.91s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX 3070 | 1.04it/s | 0.96s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| Intel A770 | 1.02it/s | 0.98s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 3060 Ti | 0.85it/s | 1.18s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3060 | 0.67it/s | 1.49s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA GTX 1080 Ti | 0.15it/s | 6.43s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux |
1024px
Runs on 2GB of VRAM with tiled VAE. 4GB can run the regular variant
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 15.02it/s | 0.0665s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 14.15it/s | 0.0707s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 5090 | 8.95it/s | 0.11s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 4090 | 7.00it/s | 0.14s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| AMD RX 7900 XTX | 4.92it/s | 0.20s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| AMD RX 7900 XTX | 4.80it/s | 0.21s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3090 | 4.00it/s | 0.25s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 2.81it/s | 0.36s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5060 Ti | 2.60it/s | 0.39s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| Intel A770 | 2.40it/s | 0.42s/it | OpenVINO 2025.2 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 3070 | 2.34it/s | 0.43s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 Ti | 1.96it/s | 0.51s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| Intel A770 | 1.75it/s | 0.56s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 3060 | 1.49it/s | 0.67s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA GTX 1080 Ti | 0.31it/s | 3.22s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA GTX 980 | 0.18it/s | 5.35s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD EPYC 9654 | 0.14it/s | 7.13s/it | CPU | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| AMD Pro W5500 | 0.13it/s | 7.35s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Arch Linux | |
| NVIDIA GTX 1050 Ti | 0.0833it/s | 12.00s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro W5500 | 0.0699it/s | 14.31s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| AMD Ryzen 5800X | 0.0365it/s | 27.42s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA MX 330 | 0.0289it/s | 34.60s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.0274it/s | 36.44s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro WX 4100 | 0.0247it/s | 40.50s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| AMD Pro W5500 | 0.0147it/s | 68.04s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| Intel i7-1165G7 | 0.0145it/s | 69.08s/it | CPU | ComfyUI (Unknown) | Arch Linux |
512px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 24.49it/s | 0.0408s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 23.42it/s | 0.0427s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| NVIDIA RTX 5090 | 21.52it/s | 0.0465s/it | CUDA 12.8 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX 4090 | 18.5it/s | 0.05s/it | CUDA 12.9 | ComfyUI (Unknown) | Windows 11 24H2 | |
| NVIDIA RTX A4000 | 15.62it/s | 0.06s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| AMD RX 7900 XTX | 14.39it/s | 0.07s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3090 | 12.39it/s | 0.0807s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 10.80it/s | 0.09s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3060 Ti | 9.54it/s | 0.10s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3070 | 8.54it/s | 0.12s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 5060 Ti | 8.53it/s | 0.12s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| Intel A770 | 7.20it/s | 0.14s/it | OpenVINO 2025.2 | SDNext (a9c65c0e) | Arch Linux | |
| AMD RX 7900 XTX | 6.60it/s | 0.15s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| NVIDIA RTX 3060 | 4.17it/s | 0.24s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| Intel A770 | 1.70it/s | 0.59s/it | Pytorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 1080 Ti | 1.20it/s | 0.83s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| AMD EPYC 9654 | 0.93it/s | 1.08s/it | CPU | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA GTX 980 | 0.69it/s | 1.45s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro W5500 | 0.54it/s | 1.85s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| AMD EPYC 9654 | 0.47it/s | 2.12s/it | CPU | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| AMD Pro W5500 | 0.42it/s | 2.38s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| NVIDIA GTX 1050 Ti | 0.29it/s | 3.40s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro W5500 | 0.20it/s | 5.06s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Arch Linux | |
| AMD Ryzen 5800X | 0.19it/s | 5.32s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.17it/s | 5.85s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA MX 330 | 0.11it/s | 9.25s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD HD 7790 | 0.11it/s | 9.39s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| AMD Pro WX 4100 | 0.1043it/s | 9.59s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| Intel i7-1165G7 | 0.0717it/s | 13.94s/it | CPU | ComfyUI (Unknown) | Arch Linux |
SD1.5
512px
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 51.83it/s | 0.019s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 46.93it/s | 0.0213s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| AMD RX 7900 XTX | 26.69it/s | 0.04s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| AMD RX 7900 XTX | 22.75it/s | 0.04s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| NVIDIA RTX 3090 | 20.58it/s | 0.0486s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX 5060 Ti | 14.21it/s | 0.0704s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 15.62it/s | 0.06s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3070 | 13.78it/s | 0.07s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| Intel A770 | 11.85it/s | 0.08s/it | OpenVINO 2025.2 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 3060 Ti | 11.58it/s | 0.09s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| Intel A770 | 6.40it/s | 0.15s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 3060 | 4.17it/s | 0.24s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA GTX 1080 Ti | 3.20it/s | 0.31s/it | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA GTX 980 | 1.59it/s | 0.63it/s | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro W5500 | 1.01it/s | 0.99s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Arch Linux | |
| AMD Pro W5500 | 0.78it/s | 1.27s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| AMD Pro W5500 | 0.75it/s | 1.32s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| NVIDIA GTX 1050 Ti | 0.71it/s | 1.40s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Ryzen 5800X3D | 0.31it/s | 3.20s/it | CPU | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA MX 330 | 0.30it/s | 3.37s/it | CUDA 12.4 | ComfyUI (Unknown) | Atch Linux | |
| AMD Pro WX 4100 | 0.24it/s | 4.07s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| AMD Pro WX 4100 | 0.22it/s | 4.38s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| AMD Ryzen 5800X | 0.22it/s | 4.73s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.21it/s | 4.87s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-1165G7 | 0.11it/s | 8.97s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| AMD RX 550 | 0.0651it/s | 15.35s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Arch Linux | 64 bit bus |
| Intel i7-3770 | 0.0569it/s | 17.57s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i5-5257U | 0.05456it/s | 18.33s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i5-6300U | 0.0295it/s | 33.85s/it | CPU | KoboldCPP (1.95.1) | Arch Linux | |
| MediaTek Dimensity 1080 | 0.0240it/s | 41.71s/it | CPU | SDNext (a9c65c0e) | HyperOS 2.0.5.0 (A14 - 4.19.191) | |
| Intel i5-4300U | 0.0117it/s | 85.34s/it | CPU | KoboldCPP (1.95.1) | Arch Linux |
256px
Runable even on 1GB of VRAM!
| Chip | it/s | s/it | Backend | App (Commit) | OS | Notes |
|---|---|---|---|---|---|---|
| NVIDIA RTX PRO 6000 B | 53.24it/s | 0.0188s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 5090 | 47.02it/s | 0.0213s/it | CUDA 12.9 | ComfyUI (37d620a) | Fedora 42 | |
| AMD RX 7900 XTX | 34.70it/s | 0.03s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | Flash Attention, OC |
| NVIDIA RTX 3090 | 33.85it/s | 0.0295s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA RTX A4000 | 31.52it/s | 0.03s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| NVIDIA RTX 3070 | 30.35it/s | 0.03s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA RTX 3060 Ti | 27.28it/s | 0.04s/it | CUDA 12.9 | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| Intel A770 | 26.30it/s | 0.04s/it | OpenVINO 2025.2 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA RTX 5060 Ti | 25.45it/s | 0.0409s/it | CUDA 12.9 | ComfyUI (Unknown) | Arch Linux | |
| AMD RX 7900 XTX | 20.51it/s | 0.05s/it | ROCm 6.4.1 | SDNext (aa0652ca) | Arch Linux | INT8 Matmul, Flash Attention, OC |
| NVIDIA RTX 3060 | 15.56it/s | 0.0643s/it | CUDA 12.9 | ComfyUI (Unknown) | Unknown | |
| NVIDIA GTX 1080 Ti | 8.23it/s | 0.12it/s | CUDA 12.6 | ComfyUI (Unknown) | Arch Linux | |
| Intel A770 | 6.20it/s | 0.16s/it | PyTorch XPU 2.7.1 | SDNext (a9c65c0e) | Arch Linux | |
| NVIDIA GTX 980 | 4.43it/s | 0.23s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro W5500 | 3.84it/s | 0.26s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Arch Linux | |
| AMD Pro W5500 | 2.84it/s | 0.35s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| AMD Pro W5500 | 2.05it/s | 0.48s/it | DirectML | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA GTX 1050 Ti | 1.82it/s | 0.55s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD EPYC 9654 | 1.78it/s | 0.56s/it | CPU | ComfyUI (Unknown) | Ubuntu Server 24.04.2 LTS | |
| AMD Ryzen 5800X3D | 1.16it/s | 0.86s/it | CPU | SDNext (a9c65c0e) | Arch Linux | |
| AMD Ryzen 5800X | 1.02it/s | 0.98s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-10700K | 0.83it/s | 1.21s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| NVIDIA MX 330 | 0.75it/s | 1.34s/it | CUDA 12.4 | ComfyUI (Unknown) | Arch Linux | |
| AMD Pro WX 4100 | 0.71it/s | 1.41s/it | Vulkan 1.3 | KoboldCPP (1.95.1) | Windows 11 22H2 | |
| AMD Pro WX 4100 | 0.66it/s | 1.50s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| AMD HD 7790 | 0.60it/s | 1.65s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| Intel i7-1165G7 | 0.48it/s | 2.07s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| AMD HD 7750 | 0.48it/s | 2.08s/it | DirectML | ComfyUI (Unknown) | Windows 11 22H2 | |
| Intel i7-4790K | 0.40it/s | 2.46s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i7-3770 | 0.26it/s | 3.84s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| Intel i5-6300U | 0.14it/s | 6.98s/it | CPU | KoboldCPP (1.95.1) | Arch Linux | |
| Intel i5-5257U | 0.23it/s | 4.31s/it | CPU | ComfyUI (Unknown) | Arch Linux | |
| MediaTek Dimensity 1080 | 0.10it/s | 9.57s/it | CPU | SDNext (a9c65c0e) | HyperOS 2.0.5.0 (A14 - 4.19.191) | |
| Intel i7-3770 | 0.0316it/s | 31.68s/it | Old CPU | KoboldCPP (1.95.1) | Arch Linux | |
| Intel Core 2 Quad Q9300 | 0.0081it/s | 123.34s/it | CPU Failsafe | KoboldCPP (1.95.1) | Arch Linux | |
| Intel Core 2 Duo T9300 | 0.0049it/s | 204.41s/it | CPU Failsafe | KoboldCPP (1.95.1) | Arch Linux |
How do I make my gens faster?
- Use simple samplers such as Euler instead of double step ones such as DPM 2M
- Lower your image sizes. SDXL can work coherently down to 384px and SD1.5 can go down to 128px.
- Use addons such as TeaCache
- Use low step LoRAs such as DMD2
- As a last resort, disable CFG by setting your CFG to 1. This will disable your negative prompt but also increase your speeds drastically. This will also severely affect your output quality. The lack of negative prompt can be circumvented with the use of NAG
- Upgrade your potato with a new GPU if all else fails.
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support