BabyVision
BabyVision - general visual reasoning benchmark. Reports Without CI and With CI (Consistency Inference) scores.
Category
Vision & Language
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for BabyVision
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen3.8-27B |
68.96%
|
18.08.2026 | Verified |
| Qwen3.6-27B |
21.17%
|
18.08.2026 | Unverified |
| Qwen3.7-Plus |
67.66%
|
18.08.2026 | Unverified |
| Opus4.6 Max |
12.60
|
18.08.2026 | Unverified |
| Qwen3.5-4B |
4.42%
|
01.02.2026 | Verified |
| Kimi K3 |
94.94%
|
18.08.2026 | Verified |
| Qwen3.5-122B-A10B |
35.84%
|
01.02.2026 | Verified |
| Qwen3.5-27B |
41.56%
|
01.02.2026 | Unverified |
| Qwen3.5-35B-A3B |
33.51%
|
01.02.2026 | Unverified |
| Claude-Sonnet-4.5 |
7.79%
|
01.02.2026 | Unverified |
| GPT-5-mini 2025-08-07 |
10.78%
|
01.02.2026 | Unverified |
| Qwen3-235B-A22B |
12.47%
|
01.02.2026 | Unverified |
| Qwen3-VL-235B-A22B |
12.47%
|
01.02.2026 | Verified |
| Kimi K2.6 |
35.32%
|
23.08.2026 | Verified |
| GPT-5.4 |
48.18%
|
23.08.2026 | Verified |
| Claude Opus 4.6 |
2.86%
|
23.08.2026 | Verified |
| Gemini 3.1 Pro |
50.65%
|
23.08.2026 | Verified |
| Kimi K2.5 |
31.04%
|
23.08.2026 | Verified |
| DeepSeek-V4.1-Flash |
100.00%
|
10.09.2026 | Verified |