Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.fenra.io/llms.txt

Use this file to discover all available pages before exploring further.

This guide covers how to send Anthropic usage data to Fenra.

​
Messages API

import Anthropic from '@anthropic-ai/sdk';

const anthropic = new Anthropic();

async function chat(messages) {
  const response = await anthropic.messages.create({
    model: 'claude-3-5-sonnet-20241022',
    max_tokens: 1024,
    messages
  });

  // Send to Fenra
  await fetch('https://ingest.fenra.io/usage/transactions', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'X-Api-Key': process.env.FENRA_API_KEY
    },
    body: JSON.stringify({
      provider: 'anthropic',
      model: response.model,
      usage: [{
        type: 'tokens',
        metrics: {
          input_tokens: response.usage.input_tokens,
          output_tokens: response.usage.output_tokens,
          total_tokens: response.usage.input_tokens + response.usage.output_tokens
        }
      }],
      context: {
        billable_customer_id: process.env.BILLABLE_CUSTOMER_ID
      }
    })
  });

  return response;
}

​
Prompt Caching

When using prompt caching, include the cache tokens: 
usage: [{
  type: 'tokens',
  metrics: {
    input_tokens: response.usage.input_tokens,
    output_tokens: response.usage.output_tokens,
    total_tokens: response.usage.input_tokens + response.usage.output_tokens,
    cache_read_input_tokens: response.usage.cache_read_input_tokens || 0,
    cache_creation_input_tokens: response.usage.cache_creation_input_tokens || 0
  }
}]

​
Extended Thinking

When using extended thinking, include thinking tokens: 
usage: [{
  type: 'tokens',
  metrics: {
    input_tokens: response.usage.input_tokens,
    output_tokens: response.usage.output_tokens,
    total_tokens: response.usage.input_tokens + response.usage.output_tokens,
    thinking_tokens: response.usage.thinking_tokens || 0
  }
}]

​
Streaming

When streaming, wait for the final message to get usage data: 
const stream = await anthropic.messages.stream({
  model: 'claude-3-5-sonnet-20241022',
  max_tokens: 1024,
  messages
});

const response = await stream.finalMessage();
// Now send to Fenra with response.usage

​
Supported Models

Fenra supports all Anthropic Claude models. Common models include:
ModelType
claude-3-5-sonnet-20241022Latest Sonnet
claude-3-5-haiku-20241022Latest Haiku
claude-3-opus-20240229Opus
claude-3-sonnet-20240229Sonnet (legacy)
claude-3-haiku-20240307Haiku (legacy)

​
Model Tiers

For batch processing, specify the tier: 
{
  provider: 'anthropic',
  model: 'claude-3-5-sonnet-20241022',
  model_tier: 'batch',  // 50% discount
  usage: [...]
}

​
Server-Side Tools

When using server-side tools like web search or code execution, track tool invocations: 
const response = await anthropic.messages.create({
  model: 'claude-3-5-sonnet-20241022',
  max_tokens: 1024,
  messages,
  tools: [{ type: 'web_search' }]
});

// Extract tool usage from response
const serverToolUse = response.usage.server_tool_use || {};
const webSearchRequests = serverToolUse.web_search_requests || 0;
const codeExecutionHours = serverToolUse.code_execution_hours || 0;

await fetch('https://ingest.fenra.io/usage/transactions', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'X-Api-Key': process.env.FENRA_API_KEY
  },
  body: JSON.stringify({
    provider: 'anthropic',
    model: response.model,
    usage: [
      {
        type: 'tokens',
        metrics: {
          input_tokens: response.usage.input_tokens,
          output_tokens: response.usage.output_tokens,
          total_tokens: response.usage.input_tokens + response.usage.output_tokens
        }
      },
      {
        type: 'requests',
        metrics: {
          count: webSearchRequests,
          request_type: 'web_search'
        }
      }
    ],
    context: {
      billable_customer_id: process.env.BILLABLE_CUSTOMER_ID
    }
  })
});

​
Next Steps