+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Prompting & Context Engineering (https://www.datasculpting.ai/inference/forumdisplay.php?fid=15)
+--- Forum: Structured Output (https://www.datasculpting.ai/inference/forumdisplay.php?fid=63)
+--- Thread: structured output failures (/showthread.php?tid=442)
structured output failures - batchscope72 - 09-23-2026
I'm trying to sanity-check my setup.
My current setup is Ollama with A6000, and I keep running into token cost.
If you've tested something similar, what did you keep and what did you stop bothering with?
Re: structured output failures - kernelhouse70 - 09-25-2026
My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals. Not glamorous advice, but repeatable baselines save a lot of time.
Re: structured output failures - circuitkite - 09-25-2026
Interesting, I hadn't tried that. That matches what I saw on 4090.
Re: structured output failures - minthall - 09-28-2026
Appreciate the detail. That matches what I saw on A6000.
Re: structured output failures - circuitworks90 - 09-28-2026
Quote:Appreciate the detail. That matches what I saw on A6000.
I hit almost the same issue last night.
Re: structured output failures - cudacrate - 09-29-2026
What worked for me was pretty simple. 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: structured output failures - plainline - 09-29-2026
subscribed, curious where this lands.
Re: structured output failures - vectorember - 09-29-2026
Thanks, that helps. I don't think your conclusion is far off.
Re: structured output failures - routecraft - 09-30-2026
Interesting, I hadn't tried that. 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.