how can I use bm25 retriever with Korean or Japanese ? I received this error, ValueError: ko not recognized. Only English stopwords as default, German, Dutch, French, Spanish, Portuguese, Italian, Russian, Swedish, Norwegian, and Chinese are currently supported. Please input a list of stopwords

likely from the stemmer ?

Can i know when chunking using default setting, which part of the nodes are transformed into embedding ?

Is it the nodegetcontent.metadata.embed ?

Is there a way to write a flexible and complex typesafe multi agent rag using llamaindex ?

Is it using this one ?
https://docs.llamaindex.ai/en/stable/module_guides/workflow/

is there a way to display eval result from llamaindex eval ?

i wanna eval retriever with node postprocessor using llamaindex evaluator, but I don't know where to put the reranker since the evaluator only accept retriever not node postprocessor, is there a retriever bundled with node postprocessor ?

Another day, another dependency conflict in python

any better alternative than graphrag in term of :

• cost
• speed
• simplicity ?

I'm really not sold on graphrag performance, seems a high level context, low level context, and summary could be done better using partitioned vectordb or hybrid search

Any plan to have llamaindex for rust ?

Why does the node metadata in my llamaindex is always in /8284/hsk2/82849 ? Especially for CJK fonts ?

How do i get to display citations and sources in llama index ?

Citations query engine doesn't seem to be enough

got error when running this:
https://github.com/run-llama/llama_index/blob/main/docs/docs/examples/cookbooks/GraphRAG_v1.ipynb

It has better typehints

i noticed this doesn't work on stable version of llamaindex https://docs.llamaindex.ai/en/stable/examples/agent/openai_agent_with_query_engine.html
where can I get OpenAiAgent class ?

tried


didn't work either

Is there a way to have the openaiagent behaving like openai assistant, as in their response consisting of tool call then message then tool call again until they see their response as good enough ? As opposed to the default in llamaindex openaiagent where the agent will call tool after tool and only send a message when they think the answer is good

i'm eyeing this part of the doc, and testing locally https://docs.llamaindex.ai/en/stable/examples/agent/agent_runner/agent_runner.html

my question is :

• how can I get the streaming of the step / response ?
• I saw the typehint, but the output is the same for task.arunstep and task.astreamstep

Also is there a small to big indexing / retriever

Why do i keep getting index error when using LLMRerank ?

is there a way to do hybrid retriever / bm25 retriever without storing and loading from filesystem (docstore / objectstore) ?

how can I incorporate this tool into react/ openai agent ?
https://docs.llamaindex.ai/en/stable/api_reference/tools/duckduckgo/?h=duck

Where to import ? do i need to install additional things first ?

so I'm using Unstructured node parser and got this error

I'm using OpenaiAgent stepwise, and I'm wondering how can I stream these original openai output that is saved in chathistory ?
Right now I can only manage to stream ToolOutput class, and ChatMessage class... ToolOutput class is kinda unusuable to be fedback into the agent

also is there a way to have somekind of middleware before and after the tool call made by the agent ? I want to publish what the bot is doing, what tool they chose

is there a way to attach user_id or thread_id or even api key to the tool call made by OpenaiAgent ?