multimodal benchmark
V* Leaderboard
A visual reasoning benchmark evaluating multimodal inference under challenging spatial and grounded tasks.
Updated Aug 17, 2026
Models7
Model coverage7
MetricScore
EvidenceB
V* Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelMA | Score96.9% | Percentile100.0% | Participants7 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score96.9% | Percentile83.3% | Participants7 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score94.7% | Percentile66.7% | Participants7 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score93.7% | Percentile50.0% | Participants7 | EvidenceC | Evaluated |
| Rank05 | ModelAC | Score93.2% | Percentile33.3% | Participants7 | EvidenceC | Evaluated |
| Rank06 | ModelAC | Score92.7% | Percentile16.7% | Participants7 | EvidenceC | Evaluated |
| Rank07 | ModelZA | Score89.0% | Percentile0.0% | Participants7 | EvidenceC | Evaluated |
V* Highlights
The leading models and scores on this benchmark.
V* Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for V*
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis v* AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01MAKimi K2.6Moonshot AI
- Score
- 96.9%
- Price
- $0.95 input / $4.0 output per 1M tokens
- Speed
- Up to 285 tok/s via Fireworks
Strengths
- Ranks #1 of 7 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures V*, not total model capability
- 02ACQwen3.6 PlusAlibaba Cloud / Qwen Team
- Score
- 96.9%
- Price
- $0.50 input / $3.0 output per 1M tokens
- Speed
- Up to 16 tok/s via Together
Strengths
- Ranks #2 of 7 compared models
- 83th percentile on this benchmark
- C evidence result
Considerations
- This result measures V*, not total model capability
- 03ACQwen3.6-27BAlibaba Cloud / Qwen Team
- Score
- 94.7%
- Price
- $0.60 input / $3.6 output per 1M tokens
- Speed
- Up to 6.1 tok/s via Novita
Strengths
- Ranks #3 of 7 compared models
- 67th percentile on this benchmark
- C evidence result
Considerations
- This result measures V*, not total model capability
- 04ACQwen3.5-27BAlibaba Cloud / Qwen Team
- Score
- 93.7%
- Price
- $0.30 input / $2.4 output per 1M tokens
- Speed
- Up to 6.7 tok/s via Novita
Strengths
- Ranks #4 of 7 compared models
- 50th percentile on this benchmark
- C evidence result
Considerations
- This result measures V*, not total model capability
- 05ACQwen3.5-122B-A10BAlibaba Cloud / Qwen Team
- Score
- 93.2%
- Price
- $0.40 input / $3.2 output per 1M tokens
Strengths
- Ranks #5 of 7 compared models
- 33th percentile on this benchmark
- C evidence result
Considerations
- This result measures V*, not total model capability
Selection summary
Best AI Models for V*
Kimi K2.6 currently leads V* with 96.9%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.
What is V*?
What V* measures and how its scores work.
A visual reasoning benchmark evaluating multimodal inference under challenging spatial and grounded tasks.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- V*
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- v-star|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about V*.
Which model scores highest on V*?
Kimi K2.6 is currently ranked first with 96.9%.
What does V* measure?
A visual reasoning benchmark evaluating multimodal inference under challenging spatial and grounded tasks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
7 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.