Hi @wind pasture. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
#speed
vagrant kindleBOT
I'm sorry, but without the actual content of your invoke.yaml file, it's difficult to provide specific recommendations. However, based on the knowledge sources provided, here are some general tips to speed up image generation in Invoke AI:
-
Adjust VRAM Cache Size: In your
invoke.yamlfile, you can adjust the VRAM cache size. A common solution is to reduce the size of your VRAM cache to 0.25. This ensures your system has enough memory to generate images. If you have more VRAM (like 16GB), you could try bumping it up to 8, but be careful not to set it too high as it could slow down the generation speed. Source (https://discord.com/channels/1020123559063990373/1181763961360756776) -
Use GPU for Generation: Ensure that your generations are happening on your GPU (if you have one). When you start the server, you'll see a log that looks like this:
INFO --> GPU device = cuda. If it says CUDA, you're good to go (if you have an NVIDIA GPU). Source (https://discord.com/channels/1020123559063990373/1181763961360756776) -
Batch Generation Speedup: You can speed up the generation process by using a technique called batch generation speedup. Instead of creating a batch of 100 images at
-s100, you can produce the same 100 images at-s10to-s30using a K-sampler, choose your favorite ones, and then run-s100on those images to polish some details. This technique is 3-8x as quick. Source (https://invoke-ai.github.io/InvokeAI/help/SAMPLER_CONVERGENCE#batch-generation-speedup)
- Learning Rate Adjustment: If you're training a model, you can adjust the learning rate. Higher values speed up training but risk overtraining, and lower values will take more steps to train a good model. The default of 0.0005 is conservative; you may wish to increase it to 0.005 to speed up training. Source (https://invoke-ai.github.io/InvokeAI/features/TRAINING#learning-rate)