806
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 18 Jul 2024
806 points (99.5% liked)
Technology
59654 readers
2745 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
But isn't ram slower then a GPU's vram? Last year people were complaining that suddenly local models were very slow on the same GPU, and it was found out it's because a new nvidia driver automatically turned on a setting of letting the GPU dump everything on the ram if it filled up, which made people trying to run bigger models very annoyed since a crash would be preferable to try again with lower settings than the increased generation time a regular RAM added.
Ram is slower than GPU VRAM, but that extreme slowdown is due to the bottleneck of the pcie bus that the data has to go through to get to the GPU.