Benchmark

TreeBench

TreeBench evaluates visual grounded reasoning, requiring models to localize and reason about fine-grained visual details.

Modality
multimodal
Categories
multimodal, reasoning, spatial_reasoning, vision
Openness
unknown
Source
llm_stats
Reported scores
2

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Seed 2.1 ProByteDance0.7112026-06-24evidence
Seed 2.1 TurboByteDance0.7112026-06-24evidence

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.