10-13-2026, 10:37 PM
Quote:Quote:Has anyone else run into this?
Did a small comparison this afternoon. On 4090 with llama.cpp, I saw basically the same behavior. The main thing that moved was latency, not quality.
|
M3 Max for local use: early impressions
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
M3 Max for local use: early impressions - by mlxgarden - 10-12-2026, 09:29 AM
Re: M3 Max for local use: early impressions - by torchshed66 - 10-12-2026, 09:58 PM
Re: M3 Max for local use: early impressions - by pixelcrate82 - 10-13-2026, 10:37 PM
Re: M3 Max for local use: early impressions - by quietarch - 10-15-2026, 11:23 AM
Re: M3 Max for local use: early impressions - by bytesupply76 - 10-15-2026, 12:01 PM
Re: M3 Max for local use: early impressions - by dockridge - 10-15-2026, 04:43 PM
Re: M3 Max for local use: early impressions - by plainember52 - 10-15-2026, 08:49 PM
Re: M3 Max for local use: early impressions - by quietarch - 10-17-2026, 09:07 AM
Re: M3 Max for local use: early impressions - by turboarch95 - 10-17-2026, 10:02 AM
|
Users browsing this thread: 1 Guest(s)