Model comparison
Perceptron Mk1 vs Claude 4 Sonnet
Capability radar, value bars, and side-by-side posture.
Perceptron Mk1
Not scored
value score
Claude 4 Sonnet
91
value score
Perceptron Mk1 ctx
—
Claude 4 Sonnet ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- Claude 4 Sonnet91
Attribute tape
| Field | Perceptron Mk1 | Claude 4 Sonnet |
|---|---|---|
| Developer | Perceptron AI | Anthropic |
| Context | — | 200K |
| Modalities | Text, Code, Image | |
| Openness | — | Proprietary Model |
| Speed | — | Fast |
| Price | — | $3/1M in · $15/1M out |
| Value score | — | 91 |
| Summary | Perceptron Mk1 is a closed-source vision-language model built for video understanding and embodied reasoning, accepting image and video inputs up to 2 FPS and returning structured outputs such as timecodes, bounding boxes, and points. It targets physical AI workloads including robotics, quality control, and sports analytics, priced at roughly 80–90% below comparable frontier multimodal models. | Balanced Claude model for coding and agents. |