LM Market Cap的更新时间线 - 新数据、工具、功能和平台里程碑。随我们发布而更新。
Alibaba's Qwen 3.5 lineup is now fully indexed with composite scores, benchmark results across MMLU, HumanEval, and SWE-bench, plus pricing data for all three parameter sizes.
Compare feature support across models in a single grid view. See at a glance which models support vision input, function calling, streaming, fine-tuning, extended thinking, and more.
Two new tools to help you evaluate models on performance dimensions beyond accuracy. Compare context window sizes (8K to 10M tokens) and response speed metrics side-by-side.
Stay on top of the latest model drops with our release timeline. Tracks major launches, version bumps, and feature additions from all providers in one chronological feed.
The main navigation now features rich mega-menu dropdowns for Tools, Categories, and Resources, making it faster to discover and jump to any page on the platform.
A comprehensive glossary defining 32 key AI terms from LLM and Transformer to LoRA, Quantization, and RAG. Every definition is written in plain language with context on why the term matters for model evaluation.
Dedicated benchmarks leaderboard covering MMLU, HumanEval, SWE-bench, GSM8K, Chatbot Arena Elo, and six more standardized evaluation suites with historical score tracking.
Monte Carlo-based forecast engine that projects where each model is likely to rank next month. Probability distributions, confidence intervals, and key factors driving predicted movement.
LM Market Cap goes live. Real-time composite rankings across coding, image generation, and video generation categories, powered by multiple signals updated hourly.
当我们发布新工具、添加信号或更新排名方法时收到通知。