Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix crashing when using 2.1 model (#2757)

Browse Source 
We now require more free memory to avoid attention slicing. 17.5% free
was not sufficient headroom in all cases, so now we require 25%.
This commit is contained in:
blessedcoolant 2023-02-22 08:03:51 +13:00 committed by GitHub
commit 9e5aa645a7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ldm/invoke/generator/diffusers_pipeline.py
View File

@ -330,7 +330,7 @@ class StableDiffusionGeneratorPipeline(StableDiffusionPipeline):
16 * \
latents.size(dim=2) * latents.size(dim=3) * latents.size(dim=2) * latents.size(dim=3) * \
bytes_per_element_needed_for_baddbmm_duplication
if max_size_required_for_baddbmm > (mem_free * 3.3 / 4.0): # 3.3 / 4.0 is from old Invoke code
if max_size_required_for_baddbmm > (mem_free * 3.0 / 4.0): # 3.3 / 4.0 is from old Invoke code
self.enable_attention_slicing(slice_size='max')
else:
self.disable_attention_slicing()