desktop apps around local AI for DeepSeek

[vectorport]

Spent some time on this today and wanted to compare notes.

Two paths both seem viable, but they break down differently once consumer GPUs becomes important.

What would you pick if you had to live with it for a few months?

[rocmriley]

Spent some time on this today and wanted to compare notes.

that fixed it for me too. At this point the real bottleneck is my patience.

[mlxgarden]

A couple questions before I guess too hard. What model, what runtime, and roughly what context size?

[vramstream]

I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit.

[deltagrid]

Thanks, that helps. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[rasterbench]

What worked for me was pretty simple. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[mlxgarden]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Llama and compare from there.

[rasterbench]

Interesting, I hadn't tried that. I ended up in about the same place using TensorRT-LLM.

[vramstream]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at embeddings.