Claude Sonnet 4.6 Review: The Workhorse That Ate the Flagship
Anthropic's mid-tier model delivers 98% of Opus performance at one-fifth the cost, with a 1M token context window and near-parity on coding and computer use benchmarks.
Benchmarks
Google Launches Gemini 3.1 Pro, Claims Top Spot on 13 of 16 Benchmarks
Google releases Gemini 3.1 Pro with dramatically improved reasoning, topping Claude Opus 4.6 and GPT-5.2 on most industry benchmarks.
This Week in AI Research: When Agents Try Science, Smarter Synthetic Data, and Protecting Your Model's Reasoning
Three papers that matter this week: a brutal benchmark for AI research agents, a feature-space approach to training data diversity, and trace rewriting to stop model theft.
Home GPU LLM Leaderboard: Best Open Source Models by VRAM Tier with Token/s Benchmarks
Rankings of the best open source LLMs you can run on home hardware - RTX 4090, RTX 3090, Apple M3/M4 Max - organized by VRAM tier with real-world token/s benchmarks and quality scores.
Do AI Benchmarks Still Matter? The Evidence for and Against Public Leaderboards
A data-driven look at benchmark contamination, leaderboard gaming, and whether public AI benchmarks can still tell us anything useful about model capabilities.
Long-Context Benchmarks Leaderboard: MRCR, RULER, and LongBench v2
Rankings of the best AI models for long-context tasks, measuring retrieval accuracy, reasoning, and comprehension across massive context windows from 128K to 10M tokens.
