AI labs cut prices – Fudzilla.com
…The cuts mark a shift for US AI outfits selling proprietary “closed” models, which have until now fought hardest on performance. More capable “open” Chinese models, which developers can freely download and…
Search
All sources
techcrunch.com
14
amd.com
9
arstechnica.com
6
developer.nvidia.com
6
engadget.com
5
theverge.com
5
cnet.com
5
fudzilla.com
5
spectrum.ieee.org
5
neowin.net
4
tomshardware.com
4
pcworld.com
4
Top stories
The U.S. wants to contain China’s AI. Silicon Valley keeps using it
…Synthetic data generated by one model routinely becomes training material for another. Frontier laboratories increasingly rely on teacher models to improve reasoning, coding, multilingual performance, and post-training alignment. In many cases…
Next Gen Networking Transport for Large Scale AI Training
…models can do large-scale AI model training with predictability, performance, and resilience across large scale AI training clusters. In this post, we will discuss the challenges of traditional Ethernet, why models…
OpenAI Launches Full-Scale Effort to Patch Open-Source Bugs as It Takes on Anthropic’s Mythos
…out that the model scores 85.6 percent on the benchmark assessment known as CyberGym, an improvement from a previous version of GPT-5.5-Cyber. The performance also beats Anthropic's…
Discussions and forums
Open-weight AI models are catching up to the frontier. The safety gap remains. | TechCrunch
…training data and then trains the model on the curated dataset. Some research suggests this can reduce hazardous biological knowledge without harming overall model performance. However, for cybersecurity, data filtering is much…
GPT-5.6 Luna: OpenAI cuts API prices by 80 percent
…OpenAI describes Luna as the most cost-efficient model in the GPT-5.6 family. Within the model series, Sol represents the highest performance, Terra a more balanced ratio of cost and…
AMD Advancing AI Live Blog: Tracking the Latest Developments in AI
…Models are performing greatly on AMD platform. Claude AI is doing more of hardware engineering, great opportunity for future collaboration with AMD. Lisa says this is the beginning of multi-year partnership…
Gemini 3.5 Flash might be fast enough for gen AI to make sense
…The next few months will be telling as Google looks at opening the Omni model up to more output types to see how it performs compared to Google’s other models. “We…
AMD and MLPerf Endpoints: Leading the Next Era of GenAI Benchmarking
…services, it offers a more realistic view of performance than closed evaluation models that are controlled by a single organization and offer limited transparency into methodology or governance. With its API-centric…
Chatbots take a back seat as new GPT-5.5 model focuses on getting work done
…OpenAI GPT-5.5 performs better in broader tests, handling complex knowledge work and operating software tools, suggesting reliability across longer, multi-step tasks. The model completes more tasks end-to-end…