hi i am getting 404 error '404 Not Found

At a glance

The community member is experiencing a 404 error when trying to access the API endpoint 'http://localhost:11434/api/chat', but the home page at 'http://localhost:11434/' is working. The community members suggest checking the Ollama documentation or GitHub repository to identify the correct API endpoint. They also discuss ways to handle timeouts when using Ollama, such as increasing the request timeout or using a custom LLM without Ollama. There is no explicitly marked answer, but the community members provide suggestions and guidance to help resolve the issue.

Useful resources

kkanhaiya lal bohra

hi i am getting 404 error '404 Not Found' for url 'http://localhost:11434/api/chat' and this link not working on browser but the home page is showing ollama is working when i open this http://localhost:11434/

11 comments

WWhiteFang_Jr

Hi,
hmm 🤔 by default ollama serves API call on this endpoint only.
Can you check maybe ollama provides swagger to identify on which endpoint it is serving the model call

kkanhaiya lal bohra

how to check this

WWhiteFang_Jr

Try checking on their github: https://github.com/ollama/ollama?tab=readme-ov-file#rest-api

kkanhaiya lal bohra

i try with langchain i am getting same error 404

kkanhaiya lal bohra

i fix it i was using venv and run command ollama run mistral ollama download mistral but now i am getting httpx.ReadTimeout: timed out it becuase i don't have hight cpu so how i can give more timeout time to query funcation or how i can set it to false?

WWhiteFang_Jr

Plain Text

from llama_index.llms.ollama import Ollama

llm = Ollama(model="llama2", request_timeout=60.0) # change the value here for timeout

kkanhaiya lal bohra

how i can use ollama on google colab? how to install it?

Plain Text

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
from llama_index.core.embeddings import resolve_embed_model
from llama_index.llms.ollama import Ollama


documents = SimpleDirectoryReader("data").load_data()

bge embedding model
Settings.embed_model = resolve_embed_model("local:BAAI/bge-small-en-v1.5")

ollama
Settings.llm = Ollama(model="mistral", request_timeout=30.0)

index = VectorStoreIndex.from_documents(
    documents, show_progress=True
)
query_engine = index.as_query_engine()
response = query_engine.query("What did the author do growing up?")
print(response)

WWhiteFang_Jr

You'll have to check on ollama github whether it can be done or not.

kkanhaiya lal bohra

can i use llm directly without ollama with transformers?

WWhiteFang_Jr

Yes, you can use custom llm to setup with choice of yours: https://docs.llamaindex.ai/en/stable/module_guides/models/llms/usage_custom/?h=custom#example-using-a-custom-llm-model-advanced

kkanhaiya lal bohra

thanks

Add a reply

Find answers from the community

hi i am getting 404 error '404 Not Found