NVIDIA and you.com Nemotron post-training
Company: CoreWeave
The claim, verbatim
CoreWeave partnered with NVIDIA and you.com to post-train NVIDIA Nemotron 3.5 Lightning on CoreWeave using RL rollouts, improving model reload latency 15x versus baseline.
Source (primary)
Introducing CoreWeave Forge | CoreWeave Blog - CoreWeave (-, news_article)
View cached copy (2026-10-01)Live source ↗Archive.org ↗
Quote: “We have partnered with NVIDIA and you.com’s engineering team to post-train NVIDIA Nemotron 3.5 Lightning on CoreWeave using RL rollouts in NeMo gym with you.com’s web search API, improving the model reload latency 15x against baseline.”
How we checked this
Checked on October 2, 2026. The cited source supports every part of this claim.
The post states CoreWeave partnered with NVIDIA and you.com to post-train Nemotron 3.5 Lightning on CoreWeave using RL rollouts. It says this improved model reload latency 15x against baseline.
Confirmed in the source:
- CoreWeave partnered with NVIDIA and you.com
- The partnership post-trained NVIDIA Nemotron 3.5 Lightning on CoreWeave
- The post-training used RL rollouts
- The work improved model reload latency 15x versus baseline
What we did: Read our cached copy of the publisher (https://www.coreweave.com/blog/coreweave-forge-turn-ai-iteration-into-compounding-improvement) in full (20,511 characters, retrieved October 1, 2026) and checked each assertion in the claim against it.
Additional evidence
confirms Introducing CoreWeave Forge | CoreWeave Blog - CoreWeave
Quote: “We have partnered with NVIDIA and you.com’s engineering team to post-train NVIDIA Nemotron 3.5 Lightning on CoreWeave using RL rollouts in NeMo gym with you.com’s web search API, improving the model reload latency 15x against baseline.”