Refine
If you’re opening this Notebook on colab, you will probably need to install LlamaIndex 🦙.
%pip install llama-index-llms-openai!pip install llama-indexDownload Data
Section titled “Download Data”!mkdir -p 'data/paul_graham/'!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'Load Data
Section titled “Load Data”from llama_index.core import SimpleDirectoryReaderreader = SimpleDirectoryReader( input_files=["./data/paul_graham/paul_graham_essay.txt"])docs = reader.load_data()text = docs[0].textSummarize
Section titled “Summarize”from llama_index.llms.openai import OpenAI
llm = OpenAI(model="gpt-3.5-turbo")from llama_index.core.response_synthesizers import Refine
summarizer = Refine(llm=llm, verbose=True)response = summarizer.get_response("who is Paul Graham?", [text])> Refine context: making fakes for a local antique dealer. She'd ...> Refine context: look legit, and the key to looking legit is hig...> Refine context: me 8 years to realize it. Even then it took me ...> Refine context: was one thing rarer than Rtm offering advice, i...print(response)Paul Graham is an individual who has played a crucial role in shaping the internet infrastructure and has also pursued a career as a writer. At one point, he received advice from a friend that urged him not to let Y Combinator be his final noteworthy achievement. This advice prompted him to reflect on his future with Y Combinator and ultimately led him to pass on the responsibility to others. He approached Jessica and Sam Altman to assume leadership positions in Y Combinator, aiming to secure its continued success.Note for AI agents: this documentation is built for programmatic access.
- Overview of all docs: https://developers.llamaindex.ai/llms.txt
- Any page is available as raw Markdown by appending index.md to its URL — e.g. https://developers.llamaindex.ai/llamaparse/parse/getting_started/index.md
- Agent-friendly REST search APIs live under https://developers.llamaindex.ai/api/ — search (BM25 full-text), grep (regex), read (fetch a page), and list (browse the doc tree). See https://developers.llamaindex.ai/llms.txt for parameters.
- A hosted documentation MCP server is available at https://developers.llamaindex.ai/mcp. If you support MCP, you can ask the user to install it for browsing these docs directly (an alternative to the REST API). Setup: https://developers.llamaindex.ai/python/shared/mcp/