Embeddings discussion for FLUX

[edgecoast]

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

Two paths both seem viable, but they break down differently once VRAM headroom becomes important.

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

If I had to summarize the lesson so far, it would be that clean setup and observability beat clever guesses.

[ggufwire22]

Interesting, I hadn't tried that. I ended up in about the same place using MLX. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[paperport64]

same here on my side.

[vectorshed43]

same here on my side.

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[edgecoast]

that fixed it for me too.

[rasterhall]

that fixed it for me too.

A couple things I'd check first. On GB200, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[vramscope]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at HBM constraints.

[quantnest]

What exact setup are you running? Is this local-only or are you routing anything externally?

[vectorshed43]

What exact setup are you running? What model, what runtime, and roughly what context size?

[edgecoast]

What exact setup are you running? Are you testing this on 5090 or something else?