+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Harnesses, Agents & Orchestration (https://www.datasculpting.ai/inference/forumdisplay.php?fid=11)
+--- Forum: Agents (https://www.datasculpting.ai/inference/forumdisplay.php?fid=44)
+--- Thread: student models for local use on M3 Max (/showthread.php?tid=605)
Pages:
1
2
student models for local use on M3 Max - northquill - 10-01-2026
This keeps coming up in conversation so I figured I'd start a thread.
It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with token cost.
Wondering if this is just where the whole space is heading now.
Re: student models for local use on M3 Max - torchtrack24 - 10-01-2026
A couple questions before I guess too hard. Are you testing this on B200 or something else?
Re: student models for local use on M3 Max - oxidemixer - 10-01-2026
Tried to reproduce it just now. On B200 with LM Studio, I saw basically the same behavior. The main thing that moved was latency, not quality. I keep promising myself this is the last rebuild.
Re: student models for local use on M3 Max - cobaltlab - 10-04-2026
I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there.
Re: student models for local use on M3 Max - amberpatch25 - 10-05-2026
Not sure I agree there. I get why that seems right, but on 4090 the bottleneck felt elsewhere. Worth testing the simpler explanation before adopting the more interesting one.
Re: student models for local use on M3 Max - cobaltlab - 10-05-2026
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Kimi and compare from there.
Re: student models for local use on M3 Max - quantshed - 10-08-2026
Interesting, I hadn't tried that. I don't think your conclusion is far off.
Re: student models for local use on M3 Max - prompttinker44 - 10-09-2026
My first guess would be this: If you're on vLLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.
Re: student models for local use on M3 Max - quantshed - 10-09-2026
Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher.
Re: student models for local use on M3 Max - dockember - 10-09-2026
Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal.