Add blog post: why day zero for Nemotron 3.5 Lightning wasn't a scramble - #35
Merged
Merged
Conversation
NVIDIA released Nemotron-3.5-Lightning today, and it needed real engine work to serve well: a hybrid Mamba-Transformer architecture, NVFP4 kernels, a state-space cache, and a new reasoning parser. Readers evaluating Modelplane have no worked example of what that means for a platform that doesn't wrap the serving engine. This adds "Why Day 0 for Nemotron 3.5 Lightning wasn't a scramble", walking through the ModelCache and ModelDeployment recipe that got the model running the same day, and arguing why that's a property of Modelplane's design (it never sits between a new engine build and running it) rather than a one-off. Includes Nic Cope's author entry and avatar. Signed-off-by: Nic Cope <nicc@rk0n.org>
The post rendered with the generated OG card as a placeholder since no cover art existed yet. This adds a designed cover (PNG + SVG source) and points the post's frontmatter at it. Signed-off-by: Nic Cope <nicc@rk0n.org>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What and why
NVIDIA released Nemotron-3.5-Lightning today, and serving it well needed real engine work: a hybrid Mamba-Transformer architecture, NVFP4 kernels, a state-space cache, and a new reasoning parser. Readers evaluating Modelplane have no worked example of what that means for a platform that doesn't wrap the serving engine.
This adds the post "Why Day 0 for Nemotron 3.5 Lightning wasn't a scramble": the ModelCache and ModelDeployment recipe that got the model running the same day it shipped, and why that's a property of Modelplane's design rather than a one-off — it never sits between a new engine build and running it, so a new model arrives on the engine's release cadence instead of Modelplane's. Includes Nic Cope's author entry and avatar; no cover image yet, so it falls back to the site's generated OG card.
Checklist
npm run lintandnpm run buildlocally. (Build passes.npm run linthas no committed ESLint config in this repo, sonext lintonly offers an interactive setup prompt — it's also not part of theblog-checkCI workflow.)npm run validate-postsif I changed blog content.git commit -s.