llama-index-callbacks-nebeso

LlamaIndex callback handler for NeBeso — open source AI cost tracking.

Track every LLM call in your LlamaIndex app: tokens, cost, latency, and custom tags — all visible in your NeBeso dashboard in real time.

Installation

pip install llama-index-callbacks-nebeso

Quick start

from llama_index_callbacks_nebeso import NeBesoHandler
from llama_index.core import Settings
from llama_index.core.callbacks import CallbackManager

handler = NeBesoHandler(
    api_key="nb_ak_...",          # or set NEBESO_API_KEY env var
    tags={"feature": "rag-search", "env": "production"},
)
Settings.callback_manager = CallbackManager([handler])

# Now build your index and query normally — all LLM calls are tracked.

Usage with RAG pipelines

from llama_index_callbacks_nebeso import NeBesoHandler
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
from llama_index.core.callbacks import CallbackManager

handler = NeBesoHandler(api_key="nb_ak_...")
Settings.callback_manager = CallbackManager([handler])

documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents)

query_engine = index.as_query_engine()
response = query_engine.query("What are the main topics?")
# Token usage and cost are now visible in your NeBeso dashboard.

Supported providers

Works with any LlamaIndex-supported LLM: OpenAI, Anthropic, Google Gemini, Mistral, Meta Llama, Cohere, AWS Bedrock, and more. Cost is calculated automatically from a built-in pricing table covering 60+ models.

Environment variables

Links

• Full docs
• LangChain integration
• NeBeso dashboard
• GitHub