LlamaIndex (gateway)

1
import os
2
from llama_index.core import Document, Settings, SummaryIndex
3
from llama_index.embeddings.openai import OpenAIEmbedding
4
from llama_index.llms.openai import OpenAI
5
from respan import Respan, workflow
6
from respan_instrumentation_llama_index import LlamaIndexInstrumentor
7
8
respan_api_key = os.environ["RESPAN_API_KEY"]
9
respan_base_url = os.getenv("RESPAN_BASE_URL", "https://api.respan.ai/api")
10
11
respan = Respan(
12
    api_key=respan_api_key,
13
    base_url=respan_base_url,
14
    instrumentations=[LlamaIndexInstrumentor()],
15
)
16
Settings.llm = OpenAI(
17
    api_key=respan_api_key,
18
    api_base=respan_base_url,
19
    model="gpt-4o-mini",
20
)
21
Settings.embed_model = OpenAIEmbedding(
22
    api_key=respan_api_key,
23
    api_base=respan_base_url,
24
    model="text-embedding-3-small",
25
)
26
27
@workflow(name="gateway_rag_query")
28
def run_query():
29
    index = SummaryIndex.from_documents([
30
        Document(text="The Respan gateway routes LlamaIndex calls to hosted models.")
31
    ])
32
    return index.as_query_engine().query("What does the gateway do?")
33
34
print(run_query())
35
respan.flush()