+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: CPUs / NPUs / Accelerators (https://www.datasculpting.ai/inference/forumdisplay.php?fid=30)
+--- Thread: do small edge boxes make sense with llama.cpp (/showthread.php?tid=1176)
Pages:
1
2
Re: do small edge boxes make sense with llama.cpp - edgekite - 10-19-2026
That matches what I was seeing. The pattern around structured output has been pretty similar for me.
Re: do small edge boxes make sense with llama.cpp - thetanest - 10-19-2026
I'd approach it in a boring order. If you're on LM Studio, I'd confirm the version first and then reduce variables until the behavior is obvious.
Re: do small edge boxes make sense with llama.cpp - visionyard23 - 10-19-2026
Ran a quick check on my box. On GB200 with LM Studio, I saw basically the same behavior. I expected the model choice to matter more than it did.
Re: do small edge boxes make sense with llama.cpp - batchfield55 - 10-19-2026
Quote:Did a small comparison this afternoon. On 7900 XTX with TensorRT-LLM, I saw basically the same behavior. I expected the model choice to matter more th
What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: do small edge boxes make sense with llama.cpp - cobaltworks25 - 10-19-2026
Quote:[quote]Did a small comparison this afternoon. On 7900 XTX with TensorRT-LLM, I saw basically the same behavior. I expected the model choice to matter
My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.
Re: do small edge boxes make sense with llama.cpp - embershed99 - 10-19-2026
I'd approach it in a boring order. If you're on Ollama, 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.