Hugging Face's Darwin-27B-ZTC-v2 Leads System One Benchmark
Hugging Face's Darwin-27B-ZTC-v2 model ranks #1 in the System One Mosaic Benchmark with a Borda score of 89.58 as of October 9, 2026.
Research
292 articles
In-depth coverage of new AI research — papers, benchmarks, and breakthroughs from leading labs and academia, summarized for fast reading and grounded in the methods that matter.
LATEST NEWS
-
-
Ethereum Researchers Warn AI Could Threaten Wallet Security
Ethereum researchers warn AI math breakthroughs could break wallet security within months, according to a post by Justin Drake.
-
Mathematicians Call for OpenAI Boycott After AI-Generated Proofs Flood Their Field
OpenAI released over 700 AI-generated mathematical papers, sparking calls for a boycott from the Association for Human Mathematics.
-
OpenAI's Math Solutions Fall Short of Field Standards
OpenAI released 719 math solutions this week, but mathematicians say they lack human understanding, with only 10 including model reasoning.
-
Intel Unveils AgentSysPerf for Agentic Stack Benchmarking
Intel announced AgentSysPerf, a benchmarking tool, on April 5, 2024, offering performance metrics for agentic systems.
-
AMD Introduces Primus Framework for Hybrid LLM Training
AMD released Primus, a framework for training hybrid large language models on Instinct GPUs, achieving final loss of 3.382 on a 300M parameter model.
-
Noitom Releases HiPHI Dataset for Humanoid Robot Learning
Noitom unveiled a 617.5-hour motion capture dataset designed to enhance humanoid robot learning, offering precise whole-body motion and human-object interaction data.
-
Zuckerberg's Biohub Launches $1.8 Billion AI Initiative to Predict Cell Behavior
Zuckerberg's Biohub has launched a $1.8 billion initiative to build AI models that predict cell behavior, with $500 million already pledged for its five-year 'Virtual Biology Initiative.'
-
Axiom Engineers Test GPT-6 Astra in Real-World Driving
Three AI engineers used GPT-6 Astra to drive a Toyota Corolla to an In-N-Out Burger, completing the task without prior training.
-
OpenAI Releases 372 AI-Generated Math Proofs on GitHub
OpenAI published 372 mathematical results generated by an internal AI model on GitHub, aiming to advance open problems in mathematics and computer science.
-
Hugging Face Details GRPO and Its Variants for LLM Alignment
Hugging Face outlines GRPO and its variants as key methods for aligning large language models, with GRPO enabling reasoning models like DeepSeek-R1.
-
OpenAI Releases Mathematical Results from Frontier Model
OpenAI released mathematical results from its frontier model on October 6, 2026, with the average result using three hours of ChatGPT Pro compute.