10-14-2026, 02:06 PM
Ran a quick check on my box. On M3 Max with PyTorch, I saw basically the same behavior.
small models can still be good
Inference
›
Prompting & Context Engineering
›
Long Context / Memory
long context that stayed useful with MLX
long context that stayed useful with MLX
|
long context that stayed useful with MLX
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
long context that stayed useful with MLX - by promptfield - 10-13-2026, 07:11 PM
Re: long context that stayed useful with MLX - by mellowsupply - 10-13-2026, 08:28 PM
Re: long context that stayed useful with MLX - by turboscope5 - 10-13-2026, 11:05 PM
Re: long context that stayed useful with MLX - by fastfield4 - 10-14-2026, 03:11 AM
Re: long context that stayed useful with MLX - by bytesupply76 - 10-14-2026, 04:04 AM
Re: long context that stayed useful with MLX - by papermixer4 - 10-14-2026, 05:01 AM
Re: long context that stayed useful with MLX - by sanescope - 10-14-2026, 01:31 PM
Re: long context that stayed useful with MLX - by southspark - 10-14-2026, 02:06 PM
|
Users browsing this thread: 1 Guest(s)