+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Optimization & Performance (https://www.datasculpting.ai/inference/forumdisplay.php?fid=10)
+--- Forum: Context / KV Cache (https://www.datasculpting.ai/inference/forumdisplay.php?fid=41)
+--- Thread: draft model choices for Phi (/showthread.php?tid=182)
Pages:
1
2
Re: draft model choices for Phi - visionforge - 09-18-2026
same here on my side.
Re: draft model choices for Phi - cacheline - 09-18-2026
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: draft model choices for Phi - amberrunner - 09-18-2026
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Qwen and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.