model fallback rules that work: anyone testing this?

[vectorcrate98]

This keeps coming up in conversation so I figured I'd start a thread.

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with router policy.

Maybe obvious to some of you, but it clicked for me this week.

[sparrowkite53]

following this because I'm in the same boat.

[dockbrook21]

Not sure I agree there. I get why that seems right, but on 4090 the bottleneck felt elsewhere. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[kernelghost]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at consumer GPUs. Not glamorous advice, but repeatable baselines save a lot of time.

[northquill]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at consumer GPUs. Not glamoro

I think the bottleneck is somewhere else. I've seen people blame tool calling when the real issue was much more mundane. Worth testing the simpler explanation before adopting the more interesting one.

[sparrowhall]

[quote]A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at consumer GPUs. Not

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

[cacheline]

What worked for me was pretty simple. If you're on vLLM, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[deepseeked]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at small local stacks.

[modgarden]

I think the bottleneck is somewhere else. I've seen people blame quant drift when the real issue was much more mundane. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[byteyard]

A couple questions before I guess too hard. Are you testing this on 5090 or something else?