Model comparison
cbook-manga-ocr vs o3
Capability radar, value bars, and side-by-side posture.
cbook-manga-ocr
Not scored
value score
o3
93
value score
cbook-manga-ocr ctx
—
o3 ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- o393
Attribute tape
| Field | cbook-manga-ocr | o3 |
|---|---|---|
| Developer | Unknown | OpenAI |
| Context | — | 200K |
| Modalities | Text, Code | |
| Openness | — | Proprietary Model |
| Speed | — | — |
| Price | — | $10/1M in · $40/1M out |
| Value score | — | 93 |
| Summary | A Japanese manga OCR model hosted on Hugging Face that extracts recognized text from manga panel image crops, with robustness for speech bubbles, vertical and horizontal text, furigana, and stylized fonts. It is derived from the open-source kha-white/manga-ocr project and tuned for manga-specific text recognition accuracy. | OpenAI reasoning model focused on hard problems. |