NVIDIA Expands Open-Weight Model Portfolio with Nemotron-3.5-Lightning
NVIDIA has released its Nemotron-3.5-Lightning-30B-A3B-BF16 model, marking a significant expansion of the company’s presence in the open-weight artificial intelligence sector. The release, hosted on Hugging Face, provides a 30-billion parameter model designed for high-performance applications.
The release has drawn attention from the developer community regarding the nature of the model’s licensing. According to discussions on Reddit, some users have noted that the Nemotron series appears to be fully open source, a distinction that separates it from other major proprietary models. Industry observers on social media suggested that the release of such models serves to increase competition among Western technology firms and provides an alternative for industries that may face restrictions when utilizing Chinese-developed models, such as those from Qwen.
While some community members have debated the utility of standardized benchmarks in predicting real-world performance, the release of the Nemotron-3.5-Lightning model continues a trend of rapid iteration among major AI developers, including Meta and NVIDIA.
Hardware Pricing Fluctuations
In a separate development affecting NVIDIA’s professional hardware segment, reports from Reddit indicate a significant price adjustment for the RTX 6000 PRO. According to unverified user reports, the price for the RTX 6000 PRO has been raised to $16,000 USD on the official NVIDIA website.
The reported price hike has prompted discussions among hardware enthusiasts regarding alternative procurement strategies. Some users have suggested that purchasing multiple Pro 5000 GPUs—which reportedly offer 72GB of RAM—may provide better value at an estimated $8,500 per unit. Additionally, some consumers have indicated they are seeking hardware through alternative suppliers, such as Alibaba, to circumvent domestic price increases.
The volatility in professional GPU pricing remains a point of contention for enterprise users and individual researchers who rely on high-end hardware for AI development and computational tasks.