CLI helpers for local models for FLUX

[mangoquill]

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?

I'm especially interested in mistakes worth avoiding.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.

[dockkite69]

Interesting, I hadn't tried that. That matches what I saw on GB200. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[relayforge]

Not sure I agree there. I've seen people blame HBM constraints when the real issue was much more mundane.

[vramstream]

I'd start with the simpler stack.

[batchshed]

Interesting, I hadn't tried that. I don't think your conclusion is far off.

[deltagrid]

Did a small comparison this afternoon. On 7900 XTX with llama.cpp, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.

[mintstation99]

Did a small comparison this afternoon. On 7900 XTX with llama.cpp, I saw basically the same behavior. The surprise was that memory pressure showed up

I'd approach it in a boring order. If you're on vLLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.

[mlxgarden]

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

[audioriver]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals. Once you know the plain path works, then add the faster path back in piece by piece.

[rocmcrate26]

Need a little more detail here. What model, what runtime, and roughly what context size?