Open Model Evaluations

Translation status

This English page provides a localized entry and navigation shell. The full article body is currently available in Chinese.

A reproducible reading series for comparing open models across task setup, version, inference conditions, cost, and deployment constraints.

Series directory

1. High-Quality Datasets in the AI Era

2. Introducing Harbor: Infrastructure for Agent Evaluation

3. Analyzing the Agent Evaluation Workflow

4. Why Evaluation Matters More Than Models in the Age of AI Agents

5. Model Capability Evaluation: From LLM Eval to Agent Eval

6. Building an Agent Eval Harness from Scratch: Task, Environment, Trial, Trajectory, and Grader

7. The Core Objects of Agent Eval: Task, Environment, Trajectory, and Grader