LangChain+Ollama+Langfuse Not Recording Token Usage
Hello,
I am using
ChatOllama
from LangChain to send LLM requests to local Ollama server. I am also integrating LangFuse with LangChain to trace the requests. The generation requests are being successfully traced, including the input and output of the model. However, the token usage is always 0. I attached a screenshot of one trace showing zero token usage.
I checked the output of LangChain's invoke method and the usage data is in the response. It's accessible via response.usage_metadata["input_tokens"]
and response.usage_metadata["output_tokens"]
.
I also tried langfuse_context.update_current_observation(usage={"input": response.usage_metadata["input_tokens"], "unit": "TOKENS"})
but it still shows zero.
I would appreciate your help on resolving this issue.Solution:Jump to solution
thanks for reporting this, can you open an issue on github for this? https://langfuse.com/issues
GitHub
Issues · langfuse/langfuse
🪢 Open source LLM engineering platform: Observability, metrics, evals, prompt management, playground, datasets. Integrates with LlamaIndex, Langchain, OpenAI SDK, LiteLLM, and more. 🍊YC W23 - Issu...
4 Replies
Solution
thanks for reporting this, can you open an issue on github for this? https://langfuse.com/issues
GitHub
Issues · langfuse/langfuse
🪢 Open source LLM engineering platform: Observability, metrics, evals, prompt management, playground, datasets. Integrates with LlamaIndex, Langchain, OpenAI SDK, LiteLLM, and more. 🍊YC W23 - Issu...
GitHub
bug: LangChain+Ollama+Langfuse Not Recording Token Usage · Issue #2...
Describe the bug I am using ChatOllama from LangChain to send LLM requests to local Ollama server. I am also integrating LangFuse with LangChain to trace the requests. The generation requests are b...
thank you
Hi, for your information, I am currently using
langfuse.generation
to update the usage tokens. It's a little bit slow because I need to fetch the whole generation and provide the start and end times to prevent messing up the latency. But it works!