when a plain Python stack is enough for FLUX

[rasterpoint]

Posting this in case it saves someone else a little time.

If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

Simple and boring is a valid answer.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.

[ggufvault]

Need a little more detail here. Are you testing this on 4090 or something else?

[rocmriley]

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

[phasespark]

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.

[vectorpoint]

Good point. That matches what I saw on B200.

[vramscope]

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

[relaysupply90]

I think the bottleneck is somewhere else. I've seen people blame NVFP4 when the real issue was much more mundane. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[nanopatch]

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

[nanobrook]

That matches what I was seeing. 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.

[vrambandit]

that fixed it for me too. The logs were somehow calmer than I was.