multimodal benchmark
TIR-Bench
A tool-calling and multimodal interaction benchmark for testing visual instruction following and execution reliability.
Updated Aug 7, 2026
Published models4
Registry coverage4
MetricScore
EvidenceB
TIR-Bench leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 61.6% | 100.0% | 4 | C | |
| 2 | AC | 59.8% | 66.7% | 4 | C | |
| 3 | AC | 55.5% | 33.3% | 4 | C | |
| 4 | AC | 53.2% | 0.0% | 4 | C |
Score distribution
Top published rows on the benchmark's original scale.
TIR-Bench highlights
The top published results on this benchmark's own scale.
What is TIR-Bench?
Definition and scoring fields from the benchmark registry.
A tool-calling and multimodal interaction benchmark for testing visual instruction following and execution reliability.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- TIR-Bench
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- tir-bench|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about TIR-Bench.
Which model scores highest on TIR-Bench?
Qwen3.6 Plus is currently ranked first with 61.6%.
What does TIR-Bench measure?
A tool-calling and multimodal interaction benchmark for testing visual instruction following and execution reliability.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
4 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.