Loading activity…
@j_dekoninck
PhD Student at ETH Zurich, working on mathematical evaluation of LLMs (MathArena, IMproofBench, BrokenMath, ...)
AI
A user says similar mistakes appeared in 1.5% of papers from the preceding two months.
AI
The test's author says four agents had 48 hours and a $500 API key, calling three results “slop.”
AI
API pricing is $2 per million input tokens and $10 per million output tokens. OpenAI reports near-Astra results on several coding and work evaluations.
AI
The announcement puts Sol only just behind GPT-6 Astra, at slightly lower cost—but says extra reasoning narrows its cost advantage.
AI
The release announcement puts GPT-6 Astra on top and says models now run in a test harness rather than directly through an API.
AI
PhD student at ETH Zurich highlights model's top ranking and efficiency on benchmarks.
