When is NVLink worth it?
A technical analysis of NVLink performance on dual RTX 3090 setups reveals that while it significantly boosts FSDP training speeds, it has negligible impact on most inference tasks. The author concludes that the high cost of NVLink bridges may not be justified for many AI workloads.
Why it matters
Understanding hardware bottlenecks is critical for AI researchers and developers optimizing expensive GPU infrastructure for large-scale model training.
Nvidia doesn’t support NVLink on consumer GPUs anymore. The last one that had it available was the RTX 3090.
Right now in April 2026, NVLink bridges are pretty expensive, $200-$400 depending on the exact size and whether you’re able to snipe one at retail. Since I have two 3090s, I wanted to test how much benefit NVLink actually provides for AI workloads.
I ran a few tests to measure the effect that adding NVLink would have on AI-related workloads:
With these, I felt I had a good mix of tests that would be affected to different degrees by inter-GPU bandwidth.
The performance on all layer split tests was unaffected. For model inference with tensor parallelism, NVLink boosted prompt processing by about 30%, while token generation speed was exactly the same.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in