Library to monitor your OpenAI usage and send metrics and logs to Grafana Cloud
Project description
OpenAI Monitoring: Monitor OpenAI API Usage with Grafana Cloud
grafana-openai-monitoring
is a Python library that provides a decorator to monitor chat completions using the OpenAI API. It facilitates sending metrics and logs to Grafana Cloud, allowing you to track and analyze OpenAI API usage and responses.
Installation
You can install grafana-openai-monitoring using pip:
pip install grafana-openai-monitoring
Usage
The following tables shows which OpenAI function correspons to which monitoing function in this library
OpenAI Function | Monitoring Function |
---|---|
ChatCompletion.create | chat_v2.monitor |
Completion.create | chat_v1.monitor |
ChatCompletions
import openai
from grafana_openai_monitoring import chat_v2
# Set your OpenAI API key
openai.api_key = "YOUR_OPENAI_API_KEY"
# Apply the custom decorator to the OpenAI API function
openai.ChatCompletion.create = chat_v2.monitor(
openai.ChatCompletion.create,
metrics_url="YOUR_PROMETHEUS_METRICS_URL", # Example: "https://prometheus.grafana.net/api/prom"
logs_url="YOUR_LOKI_LOGS_URL", # Example: "https://logs.example.com/loki/api/v1/push/"
metrics_username="YOUR_METRICS_USERNAME", # Example: "123456"
logs_username="YOUR_LOGS_USERNAME", # Example: "987654"
access_token="YOUR_ACCESS_TOKEN" # Example: "glc_eyasdansdjnaxxxxxxxxxxx"
)
# Now any call to openai.ChatCompletion.create will be automatically tracked
response = openai.ChatCompletion.create(model="gpt-4", max_tokens=1, messages=[{"role": "user", "content": "What is Grafana?"}])
print(response)
Completions
import openai
from grafana_openai_monitoring import chat_v1
# Set your OpenAI API key
openai.api_key = "YOUR_OPENAI_API_KEY"
# Apply the custom decorator to the OpenAI API function
openai.Completion.create = chat_v1.monitor(
openai.Completion.create,
metrics_url="YOUR_PROMETHEUS_METRICS_URL", # Example: "https://prometheus.grafana.net/api/prom"
logs_url="YOUR_LOKI_LOGS_URL", # Example: "https://logs.example.com/loki/api/v1/push/"
metrics_username="YOUR_METRICS_USERNAME", # Example: "123456"
logs_username="YOUR_LOGS_USERNAME", # Example: "987654"
access_token="YOUR_ACCESS_TOKEN" # Example: "glc_eyasdansdjnaxxxxxxxxxxx"
)
# Now any call to openai.Completion.create will be automatically tracked
response = openai.Completion.create(model="gpt-4", max_tokens=1, prompt="What is Grafana?")
print(response)
Configuration
To use the grafana-openai-monitoring library effectively, you need to provide the following information:
- YOUR_OPENAI_API_KEY: Replace this with your actual OpenAI API key.
- YOUR_PROMETHEUS_METRICS_URL: Replace with the URL with your Prometheus URL.
- YOUR_LOKI_LOGS_URL: Replace with the URL where you want to send Loki logs.
- YOUR_METRICS_USERNAME: Replace with the username for Prometheus.
- YOUR_LOGS_USERNAME: Replace with the username for Loki.
- YOUR_ACCESS_TOKEN: Replace with the Cloud Access Policy token required for authentication.
After configuring the parameters, the monitored API function will automatically log and track the requests and responses to the specified endpoints.
Compatibility
Python 3.7.1 and above
Dependencies
Grafana Cloud openai requests time
License
This project is licensed under the MIT License - see the LICENSE for details.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for grafana_openai_monitoring-0.0.4.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 32e479d44873d532cc8f6ef9e48a37ebab84adfd4ce4b5f16effd6beebee7813 |
|
MD5 | 4b049db0d534ae5470d29daceb0368ed |
|
BLAKE2b-256 | e77046cc2b249bc5669d3c3b9bdb5dd081684d33ad3b74bdbe8b774976e63ee5 |
Hashes for grafana_openai_monitoring-0.0.4-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6a2f3732732437c05ebcf23fe238cda855e76c46c1351317d06c68b89791bd1e |
|
MD5 | 77fbae3b14f888c5dd6c89dbbdaf2b51 |
|
BLAKE2b-256 | c9fefa0d8bb78aca6578ce68e4bb91fbec0b81c8a5950fbdc4a2ee50065c4fac |