VibeThinker-3B: A Tiny Chinese AI Model That Beats Giants at Math and Coding
Sina's VibeThinker-3B, a tiny three-billion-parameter AI model, rivals far larger systems on math and coding benchmarks but lags on factual…
Topic
Tag: AI benchmarks
Sakana AI Launches Sakana Fugu: A Model That Routes Tasks Across a Swappable Pool of LLMs
Sakana AI's new Sakana Fugu acts as an orchestration model, routing each task to the best LLM in a swappable…
OpenAI Says GPT-5.5 Cyber Beats Anthropic’s Mythos on a Security Benchmark
OpenAI's new GPT-5.5-Cyber model scored 85.6% on the CyberGym security benchmark, edging out Anthropic's Mythos 5 at 83.8%. Access is…
Sakana AI Fugu: Multi-LLM System Matches Anthropic’s Frontier Models
Sakana AI's new tool Fugu joins many AI models into one team. The company says it matches the best models,…