+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Generative AI (https://www.datasculpting.ai/inference/forumdisplay.php?fid=17)
+--- Forum: Multimodal (https://www.datasculpting.ai/inference/forumdisplay.php?fid=72)
+--- Thread: Multimodal discussion for Phi (/showthread.php?tid=630)
Pages:
1
2
Multimodal discussion for Phi - visionforge - 09-30-2026
Spent some time on this today and wanted to compare notes.
Two paths both seem viable, but they break down differently once consumer GPUs becomes important.
Trying to avoid rebuilding the whole thing twice.
Re: Multimodal discussion for Phi - visionforge - 09-30-2026
A couple things I'd check first. On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.
Re: Multimodal discussion for Phi - southtinker - 09-30-2026
Good point. I ended up in about the same place using TensorRT-LLM. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: Multimodal discussion for Phi - macrotinker - 10-03-2026
Interesting, I hadn't tried that. I ended up in about the same place using Ollama. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: Multimodal discussion for Phi - marbleyard32 - 10-05-2026
worth checking the driver first.
There are a few different conversations hiding inside this thread, so being explicit about the goal helps.
Re: Multimodal discussion for Phi - homelabhenry - 10-05-2026
I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Could be a workload difference, but I wouldn't generalize too fast from one setup.
Re: Multimodal discussion for Phi - visionforge - 10-08-2026
Quote:I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Could be a wo
I tested this tonight. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.
Re: Multimodal discussion for Phi - dockbrook21 - 10-10-2026
Quote:[quote]I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Could
What exact setup are you running? What model, what runtime, and roughly what context size?
Re: Multimodal discussion for Phi - mangovault67 - 10-10-2026
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: Multimodal discussion for Phi - macropilot - 10-13-2026
A couple things I'd check first. On Threadripper box, the easiest win is often to back off the aggressive settings and see whether the problem survives.