+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Hardware (https://www.datasculpting.ai/inference/forumdisplay.php?fid=4)
+--- Forum: GPUs (https://www.datasculpting.ai/inference/forumdisplay.php?fid=28)
+--- Thread: VRAM bottleneck on consumer cards for Qwen (/showthread.php?tid=1460)
Pages:
1
2
Re: VRAM bottleneck on consumer cards for Qwen - embersmith - 10-18-2026
Did a small comparison this afternoon. I couldn't reproduce the slowdown until I pushed context much higher. The surprise was that memory pressure showed up before throughput collapsed.
Re: VRAM bottleneck on consumer cards for Qwen - turboyard94 - 10-18-2026
Quote:Did a small comparison this afternoon. My result was mixed: first run looked bad, second run looked normal. The main thing that moved was latency, not
subscribed, curious where this lands.
Re: VRAM bottleneck on consumer cards for Qwen - mellowwire98 - 10-18-2026
What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals.
Re: VRAM bottleneck on consumer cards for Qwen - dockkite69 - 10-18-2026
My first guess would be this: Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.
Re: VRAM bottleneck on consumer cards for Qwen - paperrunner - 10-18-2026
I tested this tonight. On dual 3090 with PyTorch, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.
Re: VRAM bottleneck on consumer cards for Qwen - torchshed - 10-18-2026
Quote:My first guess would be this: Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Not glamorous advice, but
Maybe, but I had the opposite result. I've seen people blame VRAM headroom when the real issue was much more mundane. Worth testing the simpler explanation before adopting the more interesting one.
VRAM bottleneck on consumer cards for Qwen - rocmriley - 10-18-2026
Spent some time on this today and wanted to compare notes.
I'm comparing two approaches here: keep it simple on mini PC, or squeeze more out of the stack and deal with the complexity.
What would you pick if you had to live with it for a few months?
Re: VRAM bottleneck on consumer cards for Qwen - rocmriley - 10-18-2026
Quote:[quote]My first guess would be this: Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Not glamorous advic
Interesting, I hadn't tried that. I ended up in about the same place using LM Studio. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: VRAM bottleneck on consumer cards for Qwen - rocmriley - 10-19-2026
I'd approach it in a boring order. If you're on llama.cpp, 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.
Re: VRAM bottleneck on consumer cards for Qwen - fusedhall - 10-19-2026
I tested this tonight. The issue went away once I changed one setting and restarted the stack cleanly. The main thing that moved was latency, not quality.