multimodal benchmark
TempCompass Leaderboard
TempCompass is a multimodal benchmark for evaluating the temporal perception capabilities of Video Large Language Models (Video LLMs).
Updated Oct 23, 2026
Models4
Model coverage4
MetricScore
EvidenceB
TempCompass Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelBY | Score86.90% | Percentile100.00% | Participants4 | EvidenceC | Evaluated |
| Rank02 | ModelBY | Score83.70% | Percentile66.67% | Participants4 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score74.80% | Percentile33.33% | Participants4 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score71.70% | Percentile0.00% | Participants4 | EvidenceC | Evaluated |
TempCompass Highlights
The leading models and scores on this benchmark.
TempCompass Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for TempCompass
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is TempCompass?
What TempCompass measures and how its scores work.
TempCompass is a benchmark that uses conflicting videos with identical static content but differing temporal aspects to evaluate Video Large Language Models (Video LLMs).
It measures temporal perception of action, motion, speed, temporal order, and attribute changes through multi-choice QA, yes/no QA, caption matching, and caption generation, using the Score metric as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- TempCompass
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about TempCompass.
Which model scores highest on TempCompass?
Seed 1.8 is currently ranked first with 86.90%.
What are the top three models on TempCompass?
The current leaders are Seed 1.8 (86.90%), Seed 2.0 Mini (83.70%), and Qwen2.5 VL 72B Instruct (74.80%).
Which TempCompass model has the lowest official input price?
Qwen2.5 VL 7B Instruct has the lowest matched official input price at $0.35 input / $1.1 output per 1M tokens.
Which models are fastest among TempCompass results?
No matched runtime record is currently available.
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.