RAG recall vs speed tradeoff for Phi

[latentgarden]

Thought I'd post a small recipe because a lot of threads ask similar things.

Just sharing a setup that ended up working reasonably well: Kimi, Ollama, mini PC, with a lot of attention on workstation thermals.

Maybe useful as a starting point for newer folks.

[minthall]

Thought I'd post a small recipe because a lot of threads ask similar things.

My first guess would be this: If you're on LM Studio, I'd confirm the version first and then reduce variables until the behavior is obvious.

[southden]

I'd start with the simpler stack.

[stackpilot]

Good point. I ended up in about the same place using vLLM. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[signaltrack44]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tail latency. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[vectorshed43]

subscribed, curious where this lands.

[homelabhenry]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[phaseshed]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but rep

What worked for me was pretty simple. If you're on Ollama, I'd confirm the version first and then reduce variables until the behavior is obvious.

[quietsupply11]

[quote]My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice,

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

[pixelcrate82]

same here on my side.