Intel / post

How does it compare to Qwen 3.6 35B A3B, one of the most popular models on DGX…

Source: x.com
Date: Published Aug 12, 2026

Context

Generating text faster did not make Qwen 3.6 35B A3B finish tasks sooner in Alex Cheema's DGX Spark comparison. Although Qwen generated more per second in most tested settings, it produced 45% more output tokens and took 11% more steps than Nemotron 3.5 Lightning. Overall task time was 42% longer. Comparison and output histogram.

The result shows why generation speed alone can misrepresent an 's responsiveness: the amount of text and number of steps needed to finish also affect the wait.

Terms in this piece · Glossary

token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Read the source x.com