what are you all running this week

[labnest]

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

I've gone back and forth between Ollama and PyTorch for Kimi. The interesting difference for me has been MoE serving.

Interested in tradeoffs more than fan loyalty.

I care more about stability than squeezing the last possible tokens/sec out of it.

[quiettoken]

I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup. Turns out 'just one more test' was a lie.

[sanepeak48]

I'd start with the simpler stack.

[ggufwire22]

I'd start with the simpler stack.

A couple things I'd check first. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.

[thetagrid15]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Kimi and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.

[vectorshed43]

I'd approach it in a boring order. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[thetagrid15]

Appreciate the detail. I ended up in about the same place using llama.cpp. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[quiettoken]

Appreciate the detail. I ended up in about the same place using llama.cpp. I got better results once I stopped chasing one metric and just looked at w

Need a little more detail here. Is this local-only or are you routing anything externally?

[rocmrunner40]

subscribed, curious where this lands.

[plaingarden18]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at long context.