AgentBill LangChain Integration
Automatic usage tracking and billing for LangChain applications.
Installation
Install via pip:
pip install agentbill-langchain
With OpenAI support:
pip install agentbill-langchain[openai]
With Anthropic support:
pip install agentbill-langchain[anthropic]
Quick Start
from agentbill_langchain import AgentBillCallback
from langchain_openai import ChatOpenAI
from langchain.chains import LLMChain
from langchain.prompts import PromptTemplate
# 1. Initialize AgentBill callback
callback = AgentBillCallback(
api_key="agb_your_api_key_here", # Get from AgentBill dashboard
base_url="https://api.agentbill.io",
customer_id="customer-123",
debug=True
)
# 2. Create LangChain chain with callback
llm = ChatOpenAI(model="gpt-4o-mini")
prompt = PromptTemplate.from_template("Tell me a joke about {topic}")
chain = LLMChain(llm=llm, prompt=prompt)
# 3. Run - everything is auto-tracked!
result = chain.invoke(
{"topic": "programming"},
config={"callbacks": [callback]}
)
print(result["text"])
# ✅ Automatically captured:
# - Prompt text (hashed for privacy)
# - Model name (gpt-4o-mini)
# - Provider (openai)
# - Token usage (prompt + completion)
# - Latency (ms)
# - Costs (calculated automatically)
Features
- ✅ Zero-config instrumentation - Just add the callback
- ✅ Automatic token tracking - Captures all LLM calls
- ✅ Multi-provider support - OpenAI, Anthropic, any LangChain LLM
- ✅ Chain tracking - Tracks entire chain executions
- ✅ Cost calculation - Auto-calculates costs per model
- ✅ Prompt profitability - Compare costs vs revenue
- ✅ OpenTelemetry compatible - Standard observability
Advanced Usage
Track Custom Revenue
# Track revenue for profitability analysis
callback.track_revenue(
event_name="chat_completion",
revenue=0.50, # What you charged the customer
metadata={"subscription_tier": "pro"}
)
Use with Agents
from langchain.agents import initialize_agent, load_tools
tools = load_tools(["serpapi", "llm-math"], llm=llm)
agent = initialize_agent(
tools,
llm,
agent="zero-shot-react-description",
callbacks=[callback] # Add callback here
)
# All agent steps auto-tracked!
response = agent.run("What is 25% of 300?")
Use with Sequential Chains
from langchain.chains import SimpleSequentialChain
# All chain steps tracked automatically
overall_chain = SimpleSequentialChain(
chains=[chain1, chain2, chain3],
callbacks=[callback]
)
result = overall_chain.run(input_text)
Configuration
callback = AgentBillCallback(
api_key="agb_...", # Required - get from dashboard
base_url="https://...", # Required - your AgentBill instance
customer_id="customer-123", # Optional - for multi-tenant apps
account_id="account-456", # Optional - for account-level tracking
debug=True, # Optional - enable debug logging
batch_size=10, # Optional - batch signals before sending
flush_interval=5.0 # Optional - flush interval in seconds
)
How It Works
The callback hooks into LangChain's lifecycle:
- on_llm_start - Captures prompt, model, provider
- on_llm_end - Captures tokens, latency, response
- on_llm_error - Captures errors and retries
- on_chain_start - Tracks chain execution start
- on_chain_end - Tracks chain completion
All data is sent to AgentBill via the unified OTEL pipeline (otel-collector endpoint) with proper authentication.
Supported Models
Auto-cost calculation for:
- OpenAI: GPT-4, GPT-4o, GPT-3.5-turbo, etc.
- Anthropic: Claude 3.5 Sonnet, Claude 3 Opus, etc.
- Any LangChain-compatible LLM
Troubleshooting
Not seeing data in dashboard?
- Check API key is correct
- Enable
debug=Trueto see logs - Verify
base_urlmatches your instance - Check network connectivity to AgentBill
Token counts are zero?
- Some LLMs don't return token usage
- Callback will estimate based on response length
- OpenAI/Anthropic provide accurate counts
Idempotent Ingestion (X-Request-Id)
Every OTLP batch this SDK exports to /functions/v1/otel-collector carries an
X-Request-Id: <uuid4> header (auto-generated per export, reused across
transport-level retries within that export). The collector deduplicates on
(company_id, request_id), so a transient 5xx / network retry will not
double-write WAL rows, spans, or billing events.
- No configuration required — the header is emitted automatically by the v7.x line of this SDK.
- Server fallback: clients that cannot set custom headers may send
{"requestId": "..."}at the JSON envelope root. - Native OTLP exporters: the standard OTLP-HTTP exporter keeps headers
static across batches, so per-batch rotation is not yet wired on that path
(
PHASE3-R3-SDK-REQUESTID-NATIVE— tracked); the collector still dedupes those payloads on the natural key(company_id, trace_id, span_id).
See the OpenTelemetry Integration docs for the full request-id contract.
License
MIT
Metadata
Release files for agentbill-py-langchain 11.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentbill_py_langchain-11.0.2.tar.gz | 25.3 kB | Without provenance | Details for agentbill_py_langchain-11.0.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentbill_py_langchain-11.0.2-py3-none-any.whl | Python 3 | none | any | Details for agentbill_py_langchain-11.0.2-py3-none-any.whl |
Total release size: 51.6 kB
Release files / agentbill_py_langchain-11.0.2.tar.gz
| Download URL | agentbill_py_langchain-11.0.2.tar.gz |
|---|---|
| Size | 25.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
4d94136f8e8a3f23ca6672aa83532aa0c7e771167de10e0115ade13e925593fa
|
|
BLAKE2b-256 checksum How to use checksums |
23ee56bde49314ee46bea3192931d27389d2f200f5c3645f77d8ee59714cd9ad
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.15
|
Release files / agentbill_py_langchain-11.0.2-py3-none-any.whl
| Download URL | agentbill_py_langchain-11.0.2-py3-none-any.whl |
|---|---|
| Size | 26.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
23492af97ca4d74606463b875687d480cac5346d70963e173c2c0da1cb5f0e68
|
|
BLAKE2b-256 checksum How to use checksums |
cafa3ee3994f15a403c2c5aba1cdb0b5f69ac75a43e4ea58153c818bc3818973
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.11.15
|
Security for agentbill-py-langchain 11.0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.