deepseek qwen 7b abliterated has been submitted w/ the next release
@localghost
-
Running R1 Distillations on Qwen and Llama Models
By
–
you can run the distillations of r1 which are tunes on qwen and llama – full r1 needs a few hundred gb of ram
-
Self-driving technology improves continuously despite initial limitations
By
–
yeah but it’s driving itself even if it’s tailored for sf. it can always get better, and so can everything else on the list
-
Weekly Self-Driving Car Experience and Autonomous Vehicle Usage
By
–
I sit in self driving cars every week
-
Apollo Reaches Top 2 in Utilities: Offline AI Model
By
–
Apollo recently reached Top 2 in Utilities. If you don’t have it already: it lets you run an AI on your phone even without internet. An offline small model backup if you ever need it. Much more to do, thanks for the support.
-
Four Game-Changing Technologies Defining the Future
By
–
supersonic planes
satellite internet everywhere
computers that can think
self-driving cars welcome to 2025 -
Ministral 8B generates code offline on mobile at 8 tokens per second
By
–
Here’s Ministral 8B writing some code on my phone at 8 tok/s
— Aaron Ng (@localghost) 28 janvier 2025
Completely offline and using our new code blocks.
Out now in the latest version. pic.twitter.com/h8RD5drfvgHere’s Ministral 8B writing some code on my phone at 8 tok/s Completely offline and using our new code blocks. Out now in the latest version.
-
Qwen 2.5 1M Release: What You Need to Know
By
–
qwen-2.5-1m was released yesterday – what am i missing?
-
Running Your Own AI: More Options Available
By
–
There is nothing quite like the feeling of running your own AI. More great options than ever.