OpenVLMRecords: Vision-Language Model Evaluation Results from VLMEvalKit

Evaluation records generated by VLMEvalKit, reflecting the OpenVLM Leaderboard. The dataset was last updated on 2025-04 08 06:23:26 and is maintained by the author VLMEval. It contains results from evaluating various Vision-Language Models (VLMs) on different benchmarks.

Use Cases

• Analyzing model performance trends based on the evaluation records
• Comparing different Vision-Language Models (VLMs) using the leaderboard results
• Reproducing or verifying benchmark evaluations mentioned in the description

Strengths

• Records are organized with a defined naming system for browsing, as mentioned in the description
• Directly linked to the OpenVLM Leaderboard, providing a structured view of model performance

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Description metadata is limited; actual data quality requires manual inspection after download

Provenance

Source

VLMEval

Collection Method

Generated by the VLMEvalKit evaluation framework

Time Range

null

Freshness

Last updated 2025-04-08 06:23:26

Geography

null