Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Multimodal & Grounded benchmark
BabyVision with Python. Every model the catalog carries a published BabyVision w/ Python value for, ranked by that value.
A Python-assisted BabyVision evaluation for fine-grained visual perception and grounded reasoning. 9 models in the ModelChoice catalog have a published value.
Each point is a model with a published BabyVision w/ Python value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
9 models with a published BabyVision w/ Python value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Tool-augmented multimodal score |
|---|---|---|---|
| 1 | Moonshot AI | 95.4 | |
| 2 | ELGlimpse | Elorian | 95.4 |
| 3 | TMInkling 2 Small | Thinking Machines Lab | 93.7 |
What the benchmark measures and how its value is published, from the source’s own record.
BabyVision w/ Python (BabyVision with Python) is a multimodal & grounded benchmark published by Kimi K3: Open Frontier Intelligence. A Python-assisted BabyVision evaluation for fine-grained visual perception and grounded reasoning. Its evaluation set is described by the source as visual perception tasks with Python. The source rates its difficulty as fine-grained visual perception.
Each model's result is published by Kimi K3: Open Frontier Intelligence as tool-augmented multimodal score. ModelChoice prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 9 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
All leaderboards · Multimodal & Grounded capability leaderboard
| 4 | DeepSeek | 92.5 |
| 5 | Alibaba | 91.3 |
| 6 | MiniMax | 91.0 |
| 7 | TMInkling 2 Mini | Thinking Machines Lab | 88.8 |
| 8 | Moonshot AI | 85.7 |
| 9 | Alibaba | 85.6 |
Evidence key: Observed
Rows are ordered by the value Kimi K3: Open Frontier Intelligence published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.