Model comparison
meta-llama/Prompt-Guard-86M vs Claude 4 Sonnet
Capability radar, value bars, and side-by-side posture.
meta-llama/Prompt-Guard-86M
Not scored
value score
Claude 4 Sonnet
91
value score
meta-llama/Prompt-Guard-86M ctx
—
Claude 4 Sonnet ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- Claude 4 Sonnet91
Attribute tape
| Field | meta-llama/Prompt-Guard-86M | Claude 4 Sonnet |
|---|---|---|
| Developer | meta-llama | Anthropic |
| Context | — | 200K |
| Modalities | Text-classification | Text, Code, Image |
| Openness | Open Weights | Proprietary Model |
| Speed | — | Fast |
| Price | — | $3/1M in · $15/1M out |
| Value score | — | 91 |
| Summary | Imported from Hugging Face Hub (meta-llama/Prompt-Guard-86M). | Balanced Claude model for coding and agents. |