Knowledge Bases and Evaluation Organizations: FlagEval (BAAI)
Knowledge Bases and Evaluation Organizations · FlagEval (BAAI). Language and multimodal evaluation system. This source has 5 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A2–B. Admission policy: All core resources are included. Each secondary source is admitted according to its own database, methodology, versions and leaderboard structure; there is no cross-organization scoring scheme.
FlagEval (Tianping) platform user manual and evaluation portal
Explains how to browse FlagEval leaderboards, apply for evaluation, use the console and reach the language and multimodal evaluations.
Evidence use: Supports the platform's evaluation process and currently available features; specific leaderboard conclusions must be tied to an archived version.
Limitations: Some features require registration and approval. Public methodology and leaderboard maintenance status must be confirmed for each release.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24
FlagEval (Tianping) evaluation system
Introduces evaluation capabilities in language, multimodal, science, safety and other areas, and the platform's structure.
Evidence use: Supports confirming FlagEval's product scope and official entry points.
Limitations: Some evaluations require an application or login. The maintenance status of leaderboards and datasets must be confirmed against the page version.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24
FlagEval evaluation workflow
Explains the process for applying for evaluation, connecting models, running tasks and publishing results.
Evidence use: Supports judging how FlagEval results are generated and reviewed.
Limitations: Some evaluations require an application or login. The maintenance status of leaderboards and datasets must be confirmed against the page version.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24
Introduction to FlagEval Arena
Explains anonymous battles, user voting and the arena ranking mechanism.
Evidence use: Supports explaining the platform's preference evaluation results.
Limitations: Some evaluations require an application or login. The maintenance status of leaderboards and datasets must be confirmed against the page version.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24
FlagEval multimodal evaluation
Summarizes tasks, datasets and metrics for vision-language and multimodal generation.
Evidence use: Supports checking the coverage of multimodal evaluations.
Limitations: Some evaluations require an application or login. The maintenance status of leaderboards and datasets must be confirmed against the page version.
Related concepts: Model Evaluation and Benchmarks
Link last checked: 2026-09-24