ngl i think it was insider. deepseek market crash is just the spin to hide it
@localghost
-
Qwen 2.5 7B Released with 1M Context Window
By
–
Qwen 2.5 7B just dropped with a 1M context window. Here it is running at 4bit @ 11 tok/s.
— Aaron Ng (@localghost) 27 janvier 2025
Should we add it to the next update? 1M of context adds a lot of possibilities. pic.twitter.com/y4uhLfHKNSQwen 2.5 7B just dropped with a 1M context window. Here it is running at 4bit @ 11 tok/s. Should we add it to the next update? 1M of context adds a lot of possibilities.
-
DeepSeek’s 10X Efficiency Breakthrough Reshapes AI Resource Demands
By
–
Take your intelligence estimates for Project Stargate and multiply them by 10. Every AI lab now has the means to make their models 10X+ more efficient because of DeekSeek. This doesn’t mean AI and chips will be used 10X less: the 10X cost decrease means it gets used 10X more.
-
Apollo 1.0.23 Adds Multi-Model Switching in Custom Server Mode
By
–
Apollo 1.0.23 is out with some improvements to custom server mode.
— Aaron Ng (@localghost) 27 janvier 2025
You can now load a list of models on your server and switch between them from your phone.
Makes it a bit easier to host and use AI servers. pic.twitter.com/jJMoAHZe39Apollo 1.0.23 is out with some improvements to custom server mode. You can now load a list of models on your server and switch between them from your phone. Makes it a bit easier to host and use AI servers.
-
Apollo Custom Backends Now Support Reasoning Tokens
By
–
Custom backends now support reasoning tokens in the latest version of Apollo. Just change the model type in settings.
-
DeepSeek R1 7B: Powerful Reasoning AI on Local Networks
By
–
Here’s deepseek r1 7b (qwen) beaming from my laptop to my phone at 60 tok/s.
— Aaron Ng (@localghost) 25 janvier 2025
Wild that you can serve such powerful reasoning AI models to your whole network with just a few apps. pic.twitter.com/RTHnrjmzRhHere’s deepseek r1 7b (qwen) beaming from my laptop to my phone at 60 tok/s. Wild that you can serve such powerful reasoning AI models to your whole network with just a few apps.
-
Future Mobile AI Models: Shift Toward Specialized Smaller Systems
By
–
near term future for mobile models might be many small specialized small ones