MLX-Serve 26.11.2 is out
⭐️ Many ways to run Deepseek from ~80gb - 512 ram. ivanfioravanti 's RDMA code is in for the lossless quant + Massive M3 Speedups Beamsters1 SushiSeek STQ TheDavidTai 's EXL3 quants UnslothAI or GSQ-RCO GGUF Quants Thank you:
I just want to climb up step by step and reach the highest level. I want to make bean cakes! Made me smile
Use the AI of large model deepseek V4.1 to generate a collection of Liu Yifei’s wedding photos.
Here's the Deepseek 4.1 1017 EXL3 Pagoda shootout with Sushi 1.75 STQ, 2 of turboderp_ 's EXL3s (1.68 and 2.25), and the 2.81 I did.
I personally think its a tossup between Sushi and EXL3 2.25 despite 2.25 preserving the KLD difference. What's also interesting is that 1.75
You have reached the end of the archive
All of deepseek