Production-grade AI application observability SDK — monitor every LLM call automatically
Project description
ObservIQ SDK
PostHog for AI agents. Monitor every LLM call in production — automatically.
What is ObservIQ?
ObservIQ is a production-grade AI observability platform. Add 2 lines of code to any AI application and instantly get:
- ✅ Real-time trace monitoring — every LLM call recorded
- ✅ Cost tracking — per feature, per user, per model
- ✅ Latency analytics — p50, p95, p99 breakdowns
- ✅ AI-powered anomaly detection — Groq detects issues automatically
- ✅ Multi-tenant — isolate data per team with API keys
Installation
pip install observiq-sdk
Quick Start
from groq import Groq
from observiq_sdk import ObservIQ
# Setup
groq_client = Groq(api_key="your-groq-key")
oiq = ObservIQ(
api_key="oiq_your_key_here",
base_url="https://api.observiq.io" # or http://localhost:8000
)
# Wrap your AI function — that's it!
response = oiq.trace(groq_client.chat.completions.create)(
model="llama-3.3-70b-versatile",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
# Trace automatically saved to ObservIQ dashboard ✅
Supported Providers
| Provider | Status |
|---|---|
| Groq | ✅ Supported |
| OpenAI | ✅ Supported |
| Anthropic | ✅ Supported |
| Any OpenAI-compatible API | ✅ Supported |
Features
Automatic Capture
ObservIQ automatically captures:
- Model name
- Input prompt (first 1000 chars)
- Output response (first 1000 chars)
- Latency (ms)
- Token usage (prompt + completion)
- Cost (USD) — calculated per model pricing
- Status (success/error)
- Error messages
Feature Tracking
response = oiq.trace(
groq_client.chat.completions.create,
feature_name="customer_support", # track by feature
user_identifier="user_123" # track by user
)(model="llama-3.3-70b-versatile", messages=[...])
Error Tracking
Errors are automatically captured — no extra code needed:
try:
response = oiq.trace(groq_client.chat.completions.create)(...)
except Exception as e:
# ObservIQ already logged this error trace
raise
Disable in Tests
oiq = ObservIQ(
api_key="oiq_xxx",
enabled=False # No traces sent in test environment
)
Dashboard
View all your traces at observiq.io
License
MIT © 2026 ObservIQ
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
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file observiq_sdk-0.1.0.tar.gz.
File metadata
- Download URL: observiq_sdk-0.1.0.tar.gz
- Upload date:
- Size: 5.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f78cbbebff24f75f8ec460d9efaeb77dfcb150fd72227f45235e3e8ee4898b7b
|
|
| MD5 |
87353e29e82b2af6db3956003e9c3ef6
|
|
| BLAKE2b-256 |
e18ee286c32c7e2cc8574f2158fff33c1f95f90027302391d2cdb4e243e53284
|
File details
Details for the file observiq_sdk-0.1.0-py3-none-any.whl.
File metadata
- Download URL: observiq_sdk-0.1.0-py3-none-any.whl
- Upload date:
- Size: 5.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1ac8ac935d082b31cd6d7c7ae5e7c6d94d08b067bd05457cf55f5b6351b67f9c
|
|
| MD5 |
3b9e63cf90304485f5edf7404ee13563
|
|
| BLAKE2b-256 |
5ededcc517cdf3fba437b0cf1cdc123f97a31051bb2ec3644d231736f725b152
|