good hosted AI app that still respects privacy for Mistral

[signalcraft14]

I think the bottleneck is somewhere else. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[thetashed24]

Interesting, I hadn't tried that. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[cozyarch]

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. I keep promising myself this is the last rebuild.

[thetarunner]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at prefill/decode split. Once you know the plain path works, then add the faster path back in piece by piece.

[latentcraft]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Llama and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.

[audiopath66]

I don't think that's the whole story. I get why that seems right, but on 4090 the bottleneck felt elsewhere. Worth testing the simpler explanation before adopting the more interesting one.