router in front of vLLM: early impressions

[batchsize]

Interesting, I hadn't tried that. I ended up in about the same place using Ollama. The biggest change for me was simplifying the stack instead of addi

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tool calling. Once you know the plain path works, then add the faster path back in piece by piece.

[signalpeak]

subscribed, curious where this lands.

[cacheline]

Interesting, I hadn't tried that. I ended up in about the same place using Ollama. The biggest change for me was simplifying the stack instead of adding one more optimization layer.