Nvidia Nemotron Model Allegedly Unfairly Compared
GaryMarcus · x · 2026-07-15
A researcher pointed out that a recent report on the SOOFI S model artificially lowered the evaluation scores of a baseline model (the fully open-source Nemotron-3-Nano) to make its own model appear to have "frontier champion-level" performance. Comparing this with Nvidia's official original Nemotron data reveals that the reference scores in the report were drastically deflated.
Related event: OpenAI Considers Major Price Cuts Amid AI Price War(3 posts)→
More from AGI Musings
- Claude Code skill uses 10 Markdown rules to make outputs ADHD-friendly — alex_verem · 2026-07-22
- AI Power Demand Exposes US Energy Gap, Urging Shift from Scarcity to Abundance — bradneuberg · 2026-07-22
- ControlAI CEO says an international ban on superintelligence is needed to avert extinction risk — zetalyrae · 2026-07-22
- Gary Marcus says LLMs still cannot really do math on their own — GaryMarcus · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- AI may make digital work infinitely leveraged while offline life gets more human — illscience · 2026-07-22