Analytics and tooling for LLMs

Understand your costs, users and models with analytics designed specifically for LLMs — plus, caching and rate limiting out of the box.

Compatible with leading LLMs:
01
Features

Monitor, analyze, then optimize

Usage monitoring

Track and analyze usage, requests, tokens, cache, and more.

Cost analytics

Understand how much users, requests and models are costing you.

LLM optimization

Easily add remote cache, rate limits, and auto retries.
02
· Analytics

Search and filter through your data

Search by keyword

Use our powerful search function to find keywords across all of your data tables.

Filter by column

Look up keywords with exact or fuzzy match, filtering as many columns as you need.
03
· Setup

Setup with two lines of code

Update the base URL and you're done — here's an example API request with OpenAI.

1
curl --request POST \
2
     --url https://openai.flowstack.ai/v1/chat/completions \
3
     --header 'Authorization: Bearer OPENAI_API_KEY' \
4
     --header 'Flowstack-Auth: Bearer FLOWSTACK_API_KEY' \
5
     --header 'Content-Type: application/json' \
6
     --data '{
7
         "model": "gpt-3.5-turbo",
8
         "messages": [
9
             {"role": "system", "content": "Hello!"}
13
         ],
16
     }'
1
import OpenAI from "openai";
2
                    
3
const openai = new OpenAI({                   
4
   apiKey: "YOUR_OPENAI_KEY",
5
   baseURL: "https://openai.flowstack.ai/v1",
6
   defaultHeaders: {
7
       "Flowstack-Auth": "Bearer YOUR_FLOWSTACK_KEY",
8
   },
9
});
10
                    
11
const chatCompletion = await openai.chat.completions.create({
12
   messages: [{ role: "user", content: "Hello!" }],
13
   model: "gpt-3.5-turbo",
14
});
1
import openai
2
                    
3
openai.api_base = "https://openai.flowstack.ai/v1"
4
openai.api_key = "YOUR_OPENAI_KEY"
5
                    
6
response = openai.ChatCompletion.create(
7
headers={"Flowstack-Auth": "Bearer YOUR_FLOWSTACK_KEY"},
8
model="gpt-3.5-turbo",
9
messages=[{"role": "user", "content": "Hello!"}]
10
)
04
· pricing

Use Flowstack for free

Flowstack is completely free — all we ask in return is feedback to help us get better.

Free
$

0

/month
Monitor all your LLMs from one place
Start for free
Use Flowstack for free
add
Unlimited API requests
add
Analytics dashboard and data tables
add
Remote caching and rate limits
add
Priority support
Enterprise

Custom

Specialized or white-label solutions
Get in touch
Custom setups such as:
add
SLAs and enhanced security
add
Optional self-hosting on AWS, GCP or Azure
add
White-label and embedded analytics
add
Dedicated support
05
Compatibility

Monitor all your LLMs from one place

Compatible with OpenAI, Anthropic, AI21, AWS Bedrock, GCP Vertex AI, and Mistral.

Available
now
Available
now
Available
now
Available
now
Available
now
Available
now