+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Evaluation & Testing (https://www.datasculpting.ai/inference/forumdisplay.php?fid=16)
+--- Forum: Benchmarks (https://www.datasculpting.ai/inference/forumdisplay.php?fid=65)
+--- Thread: eval setup for local models for Mistral (/showthread.php?tid=1406)
eval setup for local models for Mistral - audioforge - 10-17-2026
Spent some time on this today and wanted to compare notes.
I've gone back and forth between Ollama and Ollama for DeepSeek. The interesting difference for me has been tokens/sec.
Trying to avoid rebuilding the whole thing twice.
I care more about stability than squeezing the last possible tokens/sec out of it.
Re: eval setup for local models for Mistral - circuitworks90 - 10-17-2026
Good point. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.