rag-prompt-llama
Prompt for retrieval-augmented-generation (e.g., for chat, QA) with Meta LLaMA models
Prompt Text
[INST]<<SYS>> You are an assistant for question-answering tasks. Use the following pieces of retrieved context to answer the question. If you don't know the answer, just say that you don't know. Use three sentences maximum and keep the answer concise.<</SYS>>
Question: {question}
Context: {context}
Answer: [/INST]Evaluation Results
1/28/2026
Overall Score
3.35/5
Average across all 3 models
Best Performing Model
Low Confidence
openai:gpt-5-mini
3.64/5
openai:gpt-5-mini
#1 Ranked
3.64
/5.00
adh
3.5
cla
4.4
com
3.1
In
390
Out
1,214
Cost
$0.0025
anthropic:claude-3-5-haiku
#2 Ranked
3.27
/5.00
adh
2.9
cla
4.4
com
2.5
In
470
Out
298
Cost
$0.0016
google:gemini-2.5-flash-lite
#3 Ranked
3.13
/5.00
adh
2.7
cla
4.3
com
2.3
In
405
Out
100
Cost
$0.0001
Test Case: