language benchmark
CharadesSTA Leaderboard
CharadesSTA is a benchmark dataset for temporal activity localization via language queries, with sentence temporal annotations from the Charades dataset.
Updated Sep 5, 2026
Models12
Model coverage12
MetricScore
EvidenceB
CharadesSTA Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAC | Score64.80% | Percentile100.00% | Participants12 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score63.50% | Percentile90.91% | Participants12 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score63.50% | Percentile81.82% | Participants12 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score62.80% | Percentile72.73% | Participants12 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score62.70% | Percentile63.64% | Participants12 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score61.20% | Percentile54.55% | Participants12 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score59.90% | Percentile45.45% | Participants12 | EvidenceC | Evaluated |
| Rank08 | ModelAC | Score59.00% | Percentile36.36% | Participants12 | EvidenceC | Evaluated |
| Rank09 | ModelAC | Score56.00% | Percentile27.27% | Participants12 | EvidenceC | Evaluated |
| Rank10 | ModelAC | Score55.50% | Percentile18.18% | Participants12 | EvidenceC | Evaluated |
| Rank11 | ModelAC | Score54.20% | Percentile9.09% | Participants12 | EvidenceC | Evaluated |
| Rank12 | ModelAC | Score43.60% | Percentile0.00% | Participants12 | EvidenceC | Evaluated |
CharadesSTA Highlights
The leading models and scores on this benchmark.
CharadesSTA Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for CharadesSTA
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is CharadesSTA?
What CharadesSTA measures and how its scores work.
CharadesSTA contains 12,408 training and 3,720 testing segment-sentence pairs from videos with natural language descriptions and precise temporal boundaries.
It measures the localization of activities in videos based on language queries.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- CharadesSTA
- Modality
- multimodal
- Primary category
- language
- 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 CharadesSTA.
Which model scores highest on CharadesSTA?
Qwen3 VL 235B A22B Instruct is currently ranked first with 64.80%.
What are the top three models on CharadesSTA?
The current leaders are Qwen3 VL 235B A22B Instruct (64.80%), Qwen3 VL 235B A22B Thinking (63.50%), and Qwen3 VL 30B A3B Instruct (63.50%).
Which CharadesSTA 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 CharadesSTA results?
The fastest matched records are Qwen3 VL 4B Thinking (8.77 tok/s via DeepInfra).
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.