V-Star
V* - visual agent benchmark evaluating visual search and grounding capabilities.
Category
Vision & Language
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for V-Star
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3.6-27B |
90.70%
|
22.04.2026 | Verified |
| Qwen3.5-397B-A17B |
93.47%
|
22.04.2026 | Unverified |
| Gemma4-31B-it |
21.11%
|
22.04.2026 | Unverified |
| Claude 4.5 Opus |
21.11%
|
22.04.2026 | Unverified |
| Qwen3.6-35B-A3B |
79.15%
|
22.04.2026 | Unverified |
| Qwen3.5-27B |
88.19%
|
22.04.2026 | Unverified |
| Qwen3.5-4B |
64.57%
|
01.02.2026 | Verified |
| Qwen3.5-122B-A10B |
86.93%
|
01.02.2026 | Verified |
| Claude-Sonnet-4.5 |
58.60
|
01.02.2026 | Unverified |
| Qwen3.5-35B-A3B |
85.68%
|
01.02.2026 | Unverified |
| GPT-5-mini 2025-08-07 |
32.91%
|
01.02.2026 | Unverified |
| Qwen3-235B-A22B |
68.59%
|
01.02.2026 | Unverified |
| Qwen3-VL-235B-A22B |
68.59%
|
01.02.2026 | Verified |
| Kimi K2.6 |
96.23%
|
23.08.2026 | Verified |
| GPT-5.4 |
100.00%
|
23.08.2026 | Verified |
| Claude Opus 4.6 |
69.85%
|
23.08.2026 | Verified |
| Gemini 3.1 Pro |
96.23%
|
23.08.2026 | Verified |
| Kimi K2.5 |
71.11%
|
23.08.2026 | Verified |