nirixa 1.0.5

AI Observability & Cost Intelligence — track token costs, latency, and hallucination risk

nirixa

AI Observability & Cost Intelligence — track token costs, latency, and hallucination risk for every LLM call.

pip install nirixa

Quick Start

from nirixa import NirixaClient
import openai

client = NirixaClient(api_key="nirixa-your-key")

# Wrap any LLM call — zero changes to your existing code
response = client.track(
    feature="/api/chat",
    fn=lambda: openai.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": "Hello!"}]
    )
)

# response is the original OpenAI response — unchanged
print(response.choices[0].message.content)

Module-level API

import nirixa
import openai

nirixa.init(api_key="nirixa-your-key")

response = nirixa.track(
    feature="/api/summarize",
    fn=lambda: openai.chat.completions.create(...)
)

Auto-patch (track everything automatically)

from nirixa import NirixaClient
from nirixa.middleware import patch_openai

client = NirixaClient(api_key="nirixa-your-key")
patch_openai(client, feature="/api/chat")

# All openai calls now tracked automatically — no changes needed
import openai
openai.chat.completions.create(...)

Anthropic Support

from nirixa import NirixaClient
import anthropic

client  = NirixaClient(api_key="nirixa-your-key")
claude  = anthropic.Anthropic()

response = client.track(
    feature="/api/analyze",
    model="claude-3-5-sonnet-20241022",
    fn=lambda: claude.messages.create(
        model="claude-3-5-sonnet-20241022",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Hello!"}]
    )
)

What gets tracked

Metric	Description
Token cost	Per-call USD cost by feature and model
Latency	p50/p95/p99 response times
Hallucination risk	LOW / MEDIUM / HIGH scoring
Prompt drift	Output variance over time
Error rate	Failed calls by endpoint

Dashboard

View all your data at nirixa.in

---

निरीक्षा — Observe everything.