Llama 3.1 405B

Meta · llama family

Rank

Quality Score

1300ms

Avg Response

+9%

Adaptation Gain

What Refrase Does

Here is exactly how Refrase optimizes prompts for Llama 3.1 405B, rule by rule:

Grounding rules

Refrase injects grounding constraints that tell the model to only use information from the provided context, reducing hallucination and fabricated details.

JSON reinforcement

Refrase adds explicit JSON schema hints and formatting rules so the model produces valid, parseable JSON output without extra markdown or commentary.

Before / After

See how Refrase transforms a generic prompt for Llama 3.1 405B.

Original

Extract the key information from this document. Be accurate.

Adapted for Llama 3.1 405B

Extract the key information from this document.
Only use information explicitly present in the provided document. Do not infer or fabricate details.
Return valid JSON only. No markdown fences, no commentary outside the JSON object.

Try It

Your prompt134 chars

Optimized for llama-3.1-405b

Click "Refrase It" or select a model to see the optimized prompt.