# Nvidia Nemotron 3.5 Lightning

9 Hacker News commenters recommended Nvidia Nemotron 3.5 Lightning and 4 criticized it, November 2025 to November 2026. It was recommended in 2 of 13 months.

**Verdict:** Fast local model praised for decent code and an open training pipeline; critics say it overthinks and is weaker than comparable 27B and 35B models.

## Why people recommend it

- Fast on consumer hardware (2 comments): "Nemotron 3.5 Lightning in the MXFP4 quantization absolutely flies on the Strix Halo at 65-80 t/s, but it's dumb." ([source](https://news.ycombinator.com/item?id=49355870)) "Runs fast on my Mac - ~100 tokens/second - but it's a bit of an over-thinker." ([source](https://news.ycombinator.com/item?id=49261268))
- Generally good experiences, no major problems (2 comments): "no bad experiences except for running slowly" ([source](https://news.ycombinator.com/item?id=49265220)) "the new Nemotron model is good" ([source](https://news.ycombinator.com/item?id=50036715))
- Fun, good at fanfic (2 comments): "its very fun (but sadly a worse model than 27B" ([source](https://news.ycombinator.com/item?id=49265635)) "Better than average at fanfic really" ([source](https://news.ycombinator.com/item?id=49272974))

## What people warn about

- Overthinks and second-guesses itself (2 comments): "It writes pretty good code; it does seem to second-guess itself in thinking traces" ([source](https://news.ycombinator.com/item?id=49266682)) "Runs fast on my Mac - ~100 tokens/second - but it's a bit of an over-thinker." ([source](https://news.ycombinator.com/item?id=49261268))
- Weaker than 27B and 35B models (2 comments): "we can train LORAs for 35B as well, why use this (according to benchmarks) worse model for customization" ([source](https://news.ycombinator.com/item?id=49265159)) "its very fun (but sadly a worse model than 27B" ([source](https://news.ycombinator.com/item?id=49265635))
- Fails badly on some tasks (2 comments): "are, well, terrible at this. They just couldn't get the job done at all." ([source](https://news.ycombinator.com/item?id=49266534)) "Bad, bad failure and no amount of nudging helps it see the point." ([source](https://news.ycombinator.com/item?id=49266796))

## Alternatives hackers recommend

- [Qwen 3.8 27B](https://hackerapproved.com/p/qwen-3-8-27b.md): 1173 recommend
- [Z.ai GLM](https://hackerapproved.com/p/glm-5-2.md): 1043 recommend
- [Kimi](https://hackerapproved.com/p/kimi-k3.md): 931 recommend
- [Gemma 4](https://hackerapproved.com/p/gemma-4.md): 713 recommend
- [Qwen](https://hackerapproved.com/p/qwen.md): 474 recommend

## Cite as

Hacker Approved, "Nvidia Nemotron 3.5 Lightning", https://hackerapproved.com/p/nvidia-nemotron-3-5-lightning/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).