HN
Hacker News
Top
New
Best
Ask
Show
Jobs
Comment by glintik | Hacker News Reader
Full thread
glintik
·
How much RAM browser wants to run LLM local processing?
View on HN
dchest
·
Usually around 5 GB for a 7B 4-bit quantized model.
skeeter2020
·
probably less than it needs for a few dozen open tabs based on my past profiling experiences...
kevindamm
·
How long are you willing to wait?
Reply on news.ycombinator.com