Some of them take too much ram, i put them there for iPads etc
@localghost
-
Using R1 Distills Models Without RAG Implementation
By
–
no rag, but you can use r1 distills (the smaller ones)
-
Apollo Introduces Thinking Tokens Latest Version
By
–
Thinking tokens out now in the latest version of Apollo.
-
DeepSeek R1 1.5B Achieves 4o Performance on Any Hardware
By
–
Here’s Deepseek r1 1.5B thinking through a problem — it’s comparable to 4o and Claude 3.5 Sonnet in a number of domains like math. Except…
— Aaron Ng (@localghost) 24 janvier 2025
it’s a 1.5B model…
and can run on virtually any hardware. Truly a huge efficiency leap. pic.twitter.com/CjvsCaGiU3Here’s Deepseek r1 1.5B thinking through a problem — it’s comparable to 4o and Claude 3.5 Sonnet in a number of domains like math. Except… it’s a 1.5B model… and can run on virtually any hardware. Truly a huge efficiency leap.
-
Free Local AI Models vs Paid Services: Key Differences
By
–
you can use small local models for free bc they run on your phone. if you want to use openrouter or paid services they need a key
-
New Character Chat App Testing Launch
By
–
We’re testing a new app. If you’re interested in the character chat space & want to try a new app out: shoot me a dm.
-
Apollo Ranks Top 10 Utilities as Local Models Gain Momentum
By
–
Apollo is back in the Top 10 Utilities and Top 50 overall today. Exciting to see so much interest around local models. They really are a glimpse into the future.
-
1.5B Model Version Offers Significant Speed Improvements
By
–
yep, we have the 1.5b version also; its' very fast
-
1.5B Model Balance Speed Size Mobile Deployment
By
–
definitely, 1.5b's speed & size is a good balance for mobile. worth having downloaded too
-
More Bits Enable Greater Precision in Larger AI Models
By
–
more bits = more decimal places = more precision = bigger model