# Model content vs search\_results & citations references **URL:** \ **Category:** General\ **Created:** [August 22, 2025, 8:50pm UTC](https://community.perplexity.ai/t/model-content-vs-search-results-citations-references/1165 "2025-08-22T20:50:04Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![SDCUF](https://sea1.discourse-cdn.com/flex001/user_avatar/community.perplexity.ai/sdcuf/32/369_2.png) [@SDCUF](https://community.perplexity.ai/u/SDCUF)\ **Post date:** [August 22, 2025, 8:50pm UTC](https://community.perplexity.ai/t/model-content-vs-search-results-citations-references/1165/1 "2025-08-22T20:50:04Z")
Here is my proof of concept code – ```auto import asyncio import os from dotenv import load_dotenv from openai import AsyncOpenAI async def main(): client = AsyncOpenAI( api_key=os.getenv("PERPLEXITY_API_KEY"), base_url="https://api.perplexity.ai" ) response = await client.chat.completions.create( model="sonar-pro", messages=[ {"role": "system", "content": "Be precise and concise."}, {"role": "user", "content": "How many stars are there in our galaxy?"} ], stream=True ) async for event in response: print(event) if __name__ == " __main__": load_dotenv(".env.local") asyncio.run(main()) ``` Last event in the stream – (newlines inserted for read-ability) ```auto ChatCompletionChunk( id='d14cd700-0e62-4374-9583-c191d5a110f4', choices=[ Choice( delta=ChoiceDelta( content='[4].', function_call=None, refusal=None, role='assistant', tool_calls=None ), finish_reason=None, index=0, logprobs=None, message={ 'role': 'assistant', 'content': 'The best current estimates suggest the Milky Way contains **between 200 billion and 400 billion stars** , with 200 billion being a commonly cited figure but the total possibly as high as 400 billion due to uncertainties involving faint, low-mass red dwarf stars[1][2][3][4].\n\nThis uncertainty primarily arises from the difficulty in detecting very dim, low-mass stars—especially red dwarfs—which are extremely abundant but hard to observe directly[1][3]. Older estimates placed the number closer to 100 billion, but newer data from astronomical surveys like ESA’s Gaia mission suggest that number is likely too low and has been revised upward[1]. The exact count remains elusive, as ongoing studies continue to resolve these uncertainties.\n\nIn summary, while the precise number of stars in the Milky Way is not known for certain, **current scientific consensus places the count probably between 200–400 billion** [1][2][3][4].' } ) ], created=1755887465, model='sonar-pro', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage( completion_tokens=200, prompt_tokens=14, total_tokens=214, completion_tokens_details=None, prompt_tokens_details=None, search_context_size='low' ), citations=[ 'https://bigthink.com/starts-with-a-bang/how-many-stars-milky-way/', 'https://www.youtube.com/watch?v=MTeyypxmbK0', 'https://www.youtube.com/watch?v=hiWacnGH5II', 'https://en.wikipedia.org/wiki/Milky_Way', 'https://imagine.gsfc.nasa.gov/science/objects/milkyway1.html' ], search_results=[ { 'title': "Why we still don't know how many stars are in the Milky Way", 'url': 'https://bigthink.com/starts-with-a-bang/how-many-stars-milky-way/', 'date': '2025-04-02', 'last_updated': '2025-06-16' }, { 'title': 'How Many Stars Are in the Universe!?!?', 'url': 'https://www.youtube.com/watch?v=MTeyypxmbK0', 'date': '2025-05-24', 'last_updated': '2025-08-19' }, { 'title': 'How Many STARS Are In The MILKY WAY?', 'url': 'https://www.youtube.com/watch?v=hiWacnGH5II', 'date': '2024-03-22', 'last_updated': '2025-07-01' }, { 'title': 'Milky Way', 'url': 'https://en.wikipedia.org/wiki/Milky_Way', 'date': '2001-09-15', 'last_updated': '2025-08-18' }, { 'title': 'Milky Way Galaxy - Imagine the Universe! - NASA', 'url': 'https://imagine.gsfc.nasa.gov/science/objects/milkyway1.html', 'date': '2015-07-22', 'last_updated': '2025-08-11' } ] ) ChatCompletionChunk( id='d14cd700-0e62-4374-9583-c191d5a110f4', choices=[ Choice( delta=ChoiceDelta( content='', function_call=None, refusal=None, role='assistant', tool_calls=None ), finish_reason='stop', index=0, logprobs=None, message={ 'role': 'assistant', 'content': """The best current estimates suggest the Milky Way contains **between 200 billion and 400 billion stars** , with 200 billion being a commonly cited figure but the total possibly as high as 400 billion due to uncertainties involving faint, low-mass red dwarf stars[1][2][3][4]. This uncertainty primarily arises from the difficulty in detecting very dim, low-mass stars—especially red dwarfs—which are extremely abundant but hard to observe directly[1][3]. Older estimates placed the number closer to 100 billion, but newer data from astronomical surveys like ESA’s Gaia mission suggest that number is likely too low and has been revised upward[1]. The exact count remains elusive, as ongoing studies continue to resolve these uncertainties. In summary, while the precise number of stars in the Milky Way is not known for certain, **current scientific consensus places the count probably between 200–400 billion** [1][2][3][4].""" } ) ], created=1755887465, model='sonar-pro', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage( completion_tokens=200, prompt_tokens=14, total_tokens=214, completion_tokens_details=None, prompt_tokens_details=None, search_context_size='low', cost={ 'input_tokens_cost': 0.0, 'output_tokens_cost': 0.003, 'request_cost': 0.006, 'total_cost': 0.009 } ), citations=[ 'https://bigthink.com/starts-with-a-bang/how-many-stars-milky-way/', 'https://www.youtube.com/watch?v=MTeyypxmbK0', 'https://www.youtube.com/watch?v=hiWacnGH5II', 'https://en.wikipedia.org/wiki/Milky_Way', 'https://imagine.gsfc.nasa.gov/science/objects/milkyway1.html' ], search_results=[{ 'title': "Why we still don't know how many stars are in the Milky Way", 'url': 'https://bigthink.com/starts-with-a-bang/how-many-stars-milky-way/', 'date': '2025-04-02', 'last_updated': '2025-06-16' }, { 'title': 'How Many Stars Are in the Universe!?!?', 'url': 'https://www.youtube.com/watch?v=MTeyypxmbK0', 'date': '2025-05-24', 'last_updated': '2025-08-19' }, { 'title': 'How Many STARS Are In The MILKY WAY?', 'url': 'https://www.youtube.com/watch?v=hiWacnGH5II', 'date': '2024-03-22', 'last_updated': '2025-07-01' }, { 'title': 'Milky Way', 'url': 'https://en.wikipedia.org/wiki/Milky_Way', 'date': '2001-09-15', 'last_updated': '2025-08-18' }, { 'title': 'Milky Way Galaxy - Imagine the Universe! - NASA', 'url': 'https://imagine.gsfc.nasa.gov/science/objects/milkyway1.html', 'date': '2015-07-22', 'last_updated': '2025-08-11' }] ) ``` The search results and citations both have a list of items. The model response has 5 citations. From another thread it looks like this a known issue. **Questions** 1. Is it safe to assume that the search results and citations are always the same set and in the same order? 2. Are these lists 1-indexed when referenced by the model. e.g. The model’s `[1]` reference in the `content` is referring to `search_results[0]`and `citations[0]`? I couldn’t find clear documentation on these items Edit: It would be nice if Perplexity provided a customized Python type that matches the API return. Right now I’m having to roll my own based on openai.types.chat.ChatCompletionChunk ---
**Author:** ![Kesku](https://sea1.discourse-cdn.com/flex001/user_avatar/community.perplexity.ai/kesku/32/898_2.png) [@Kesku](https://community.perplexity.ai/u/Kesku)\ **Post date:** [April 7, 2026, 2:41pm UTC](https://community.perplexity.ai/t/model-content-vs-search-results-citations-references/1165/2 "2026-04-07T14:41:46Z")
Hey @SDCUF — good question. In streaming mode, the `citations` and `search_results` fields arrive in the **final chunk** of the stream, not in every chunk. Here is how it works: - During streaming, `delta.content` gives you the text incrementally with inline citation markers like `[1]`, `[2]`, etc. - The `citations` array (list of URLs) appears in the **last chunk** where `finish_reason` is set to `"stop"`. - The inline `[1]` markers correspond to indices in that `citations` array. To properly collect citations while streaming: ```python from perplexity import Perplexity client = Perplexity() stream = client.chat.completions.create( model="sonar-pro", messages=[{"role": "user", "content": "How many stars in our galaxy?"}], stream=True ) full_text = "" citations = [] for chunk in stream: if chunk.choices[0].delta.content: full_text += chunk.choices[0].delta.content # Citations appear in the final chunk if hasattr(chunk, "citations") and chunk.citations: citations = chunk.citations print(full_text) print("Sources:", citations) ``` Since you are using the OpenAI SDK, the `citations` field will not be typed on the response object — you will need to access it from the raw response dict or switch to the [Perplexity SDK](https://docs.perplexity.ai/docs/getting-started/quickstart) which has proper typing for these fields. Alternatively, the **Agent API** returns citations as annotations on the output, which can be easier to work with: ```python response = client.responses.create( preset="pro-search", input="How many stars in our galaxy?" ) # Citations are in response.output[].content[].annotations ``` Docs: [Sonar Streaming](https://docs.perplexity.ai/docs/sonar/quickstart) | [Agent API](https://docs.perplexity.ai/docs/agent-api/quickstart)