NVIDIA announced Nemotron for Science on 16 November 2026 at the SC26 supercomputing conference: a science-focused model family (Nemotron 3.5 Super VL) plus LitmusBench v0.2, a benchmark of roughly 50,000 chemistry questions. Early reported results: Lila Sciences nearly doubled quantum-dot prediction accuracy to 61% using it, and a KTH agent fixed 81% of broken molecular-dynamics test cases.
Key facts
| Item | Detail |
|---|---|
| Announced | 16 November 2026 |
| Event | SC26 supercomputing conference |
| Model family | Nemotron 3.5 Super VL (science-focused) |
| Benchmark | LitmusBench v0.2 — roughly 50,000 chemistry questions |
| Lila Sciences | Nearly doubled quantum-dot prediction accuracy to 61% using it |
| KTH agent | Fixed 81% of broken molecular-dynamics test cases |
What this means / what to watch
Nemotron for Science pairs a science-focused model family with a large chemistry benchmark — LitmusBench v0.2, with roughly 50,000 questions. The two early results reported so far are the Lila Sciences quantum-dot figure (61%) and the KTH agent's molecular-dynamics fix rate (81%). Beyond the announcement and these early results, the sources report no further detail.
A note on verification
The facts above are verified from the sources named in the research file: an AI news roundup (robfalconer) and Medium supercomputing coverage. The sourcing for this item is thin, so this post states only what those sources report and adds nothing beyond it.
FAQs
What is Nemotron for Science?
Nemotron for Science is a science-focused model family (Nemotron 3.5 Super VL) announced by NVIDIA on 16 November 2026 at the SC26 supercomputing conference, together with LitmusBench v0.2.
What is LitmusBench v0.2?
LitmusBench v0.2 is a benchmark of roughly 50,000 chemistry questions, announced alongside Nemotron for Science.
What did Lila Sciences report?
Lila Sciences nearly doubled quantum-dot prediction accuracy to 61% using Nemotron for Science.
What did the KTH agent achieve?
A KTH agent fixed 81% of broken molecular-dynamics test cases.
Source: AI news roundup (robfalconer); Medium (supercomputing coverage). Last checked: 21 November 2026.
Comments (0)
Share your question or experience about this syllabus. Your comment will appear after review.
No comments yet — be the first to share your thoughts!