Benchmark
BabyVision
Multimodal vision benchmark. Temperature=1.0, top_p=0.95, max context 164K tokens. Images resized to shorter side at least 1.5K pixels.
- Released
- 2025-12-01
- Categories
- multimodal
- Openness
- unknown
- Source
- model_reports
- Reported scores
- 1
Reported scores
model_reports
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| GLM-5.3-Flash | Z.ai | 53.4 | 2026-08-27 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.