Try AxonFlow — Free Trial Server

Try AxonFlow instantly at try.getaxonflow.com — no Docker, no installation, no license required. Register a tenant in 30 seconds and start testing governance policies with any SDK.

Quick Start

1. Register a tenant

curl -X POST https://try.getaxonflow.com/api/v1/register \
  -H "Content-Type: application/json" \
  -d '{"label":"my-trial"}'

Response:

{
  "tenant_id": "cs_7f3a9b2e-4d1c-4a8f-b9e3-1f2d3c4e5f6a",
  "secret": "a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6",
  "secret_prefix": "a1b2c3d4",
  "expires_at": "2026-05-09T13:17:40Z",
  "endpoint": "https://try.getaxonflow.com",
  "note": "This is a shared trial server. No SLA, no security guarantee..."
}

Save the secret — it is shown only once. The secret_prefix shows the first 8 characters for identification.

2. Configure your SDK

Set AXONFLOW_TRY=1 to auto-connect, or configure the endpoint manually:

Python

from axonflow import AxonFlow

# Use credentials from step 1
client = AxonFlow(
    endpoint="https://try.getaxonflow.com",
    client_id="cs_your-tenant-id",
    client_secret="your-secret",
)

Or auto-connect with environment variables:

export AXONFLOW_TRY=1
export AXONFLOW_CLIENT_ID=cs_your-tenant-id
export AXONFLOW_CLIENT_SECRET=your-secret

# With AXONFLOW_TRY=1 set, no endpoint needed
client = AxonFlow(
    client_id="cs_your-tenant-id",
    client_secret="your-secret",
)

Go

client := axonflow.NewClient(axonflow.AxonFlowConfig{
    Endpoint:     "https://try.getaxonflow.com",
    ClientID:     "cs_your-tenant-id",
    ClientSecret: "your-secret",
})

Or with AXONFLOW_TRY=1:

// With AXONFLOW_TRY=1 set, endpoint auto-connects to try.getaxonflow.com
client := axonflow.NewClient(axonflow.AxonFlowConfig{
    ClientID:     os.Getenv("AXONFLOW_CLIENT_ID"),
    ClientSecret: os.Getenv("AXONFLOW_CLIENT_SECRET"),
})

TypeScript

import { AxonFlow } from '@axonflow/sdk';

const client = new AxonFlow({
  endpoint: 'https://try.getaxonflow.com',
  clientId: 'cs_your-tenant-id',
  clientSecret: 'your-secret',
});

Or with AXONFLOW_TRY=1:

// With AXONFLOW_TRY=1 set, endpoint auto-connects to try.getaxonflow.com
const client = new AxonFlow({
  clientId: process.env.AXONFLOW_CLIENT_ID,
  clientSecret: process.env.AXONFLOW_CLIENT_SECRET,
});

Java

AxonFlow client = AxonFlow.create(AxonFlowConfig.builder()
    .endpoint("https://try.getaxonflow.com")
    .clientId("cs_your-tenant-id")
    .clientSecret("your-secret")
    .build());

Or with AXONFLOW_TRY=1:

// With AXONFLOW_TRY=1 set, endpoint auto-connects to try.getaxonflow.com
AxonFlow client = AxonFlow.create(AxonFlowConfig.builder()
    .clientId(System.getenv("AXONFLOW_CLIENT_ID"))
    .clientSecret(System.getenv("AXONFLOW_CLIENT_SECRET"))
    .build());

3. Send your first request

Python

import asyncio

async def main():
    result = await client.proxy_llm_call(
        "", "What is the capital of France?", "chat"
    )
    print(f"Success: {result.success}")

asyncio.run(main())

Go

result, err := client.ProxyLLMCall("", "What is the capital of France?", "chat", nil)
if err != nil {
    log.Fatal(err)
}
fmt.Println("Success:", result.Success)

TypeScript

const result = await client.proxyLLMCall('', 'What is the capital of France?', 'chat');
console.log('Success:', result.success);

Java

ClientResponse result = client.proxyLLMCall(ClientRequest.builder()
    .query("What is the capital of France?")
    .requestType(RequestType.CHAT)
    .build());
System.out.println("Success: " + result.isSuccess());

Registration

Endpoint: POST /api/v1/register

Field	Type	Required	Description
label	string	No	Human-readable name for your registration (max 255 chars)

The response contains your credentials. The secret field is shown only once — store it securely.

Credentials expire after 30 days. Register again to get new ones.

Rate Limits

Limit	Value	Scope
Requests per minute	20	Per tenant
Requests per day	500	Per tenant
Registrations per hour	5	Per IP address

When a limit is reached, the server returns HTTP 429 with a Retry-After header.

Available LLM Models

The trial server runs Ollama with the following models:

Model	Size	Best for
llama3.2:latest	2.0 GB	General chat, policy testing

Responses from local models are slower than cloud LLMs. Allow up to 30 seconds per request.

Want to test with your own API keys? Deploy AxonFlow self-hosted and configure any supported LLM provider.

Limitations and Disclaimers

Important — Read Before Using

try.getaxonflow.com is a shared trial server. It is provided free of charge for trial purposes only.

• No SLA — No uptime guarantee. The server may be unavailable without notice.
• No security guarantee — Do not send real PII, production data, or sensitive information.
• 30-day data retention — All data (registrations, audit logs, policies) is automatically purged after 30 days.
• Rate limited — 20 requests/minute, 500 requests/day per tenant, 5 registrations/hour per IP.
• Ollama only — Local LLM models only. No cloud LLM providers.
• AxonFlow reserves the right to wipe the instance at any time.

For production use, deploy self-hosted or contact [email protected] for enterprise licensing.

What's Next

• Self-hosted quickstart — Deploy AxonFlow locally with Docker Compose in 5 minutes
• Enterprise evaluation — Request an evaluation license for full features
• Regulated environments — Evaluation guide for regulated industries
• SDK documentation — Python | Go | TypeScript | Java

FAQ

Can I use my own API keys?

Not on try.getaxonflow.com — it runs Ollama only. Deploy self-hosted to use OpenAI, Anthropic, Azure, Gemini, Mistral, or Bedrock.

How do I migrate to self-hosted?

1. Deploy AxonFlow locally: docker compose up -d
2. Change your SDK endpoint from try.getaxonflow.com to localhost:8080
3. Your policies and configurations transfer — only the endpoint changes

What happens when my tenant expires?

After 30 days, your credentials stop working (HTTP 401). Register again to get new credentials. Old data is purged.

Is my data shared with other tenants?

No. Each tenant has a unique UUID (cs_ prefix) and all data is partitioned by tenant. Other tenants cannot see your policies, audit logs, or query history. The underlying infrastructure is shared, so do not send sensitive data.