speculative decoding in llama.cpp

[deepseeked]

Not sure I agree there. I get why that seems right, but on A6000 the bottleneck felt elsewhere.

[latentcraft]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at embeddings. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[northcraft90]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

[phasescope]

I hit almost the same issue last night.

[pixelcrate82]

Can you fill in a bit more context? What model, what runtime, and roughly what context size?

[mangotinker73]

Good point. That matches what I saw on dual 3090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.