Model comparison
Lift vs Claude 4 Sonnet
Capability radar, value bars, and side-by-side posture.
Lift
Not scored
value score
Claude 4 Sonnet
91
value score
Lift ctx
—
Claude 4 Sonnet ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- Claude 4 Sonnet91
Attribute tape
| Field | Lift | Claude 4 Sonnet |
|---|---|---|
| Developer | KCB1 | Anthropic |
| Context | — | 200K |
| Modalities | Text, Code, Image | |
| Openness | — | Proprietary Model |
| Speed | — | Fast |
| Price | — | $3/1M in · $15/1M out |
| Value score | — | 91 |
| Summary | Lift is a vision-language model designed to parse complex documents—combining text and images—and reconstruct their structure (headings, tables, figures, formulas, reading order) into machine-readable, structured outputs. It supports GGUF quantization for efficient local inference via llama.cpp. | Balanced Claude model for coding and agents. |