NullSpend

Get cost tracking for your Anthropic calls in under 2 minutes.

Prerequisites

A NullSpend account (sign up)
An existing app that calls the Anthropic API

Step 1: Create an API Key

Log in to the NullSpend dashboard
Go to Settings → Create API Key
Copy the key (starts with ns_live_sk_) — you won't see it again

Step 2: Point Your SDK at the Proxy

Set two environment variables:

# Point Anthropic SDK at NullSpend
ANTHROPIC_BASE_URL=https://proxy.nullspend.dev

# Your real Anthropic key — unchanged
ANTHROPIC_API_KEY=sk-ant-your-anthropic-key

# NullSpend API key
NULLSPEND_API_KEY=ns_live_sk_your-key-here

Then add the X-NullSpend-Key header to your client:

TypeScript

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

const anthropic = new Anthropic({
  baseURL: "https://proxy.nullspend.dev",
  defaultHeaders: {
    "X-NullSpend-Key": process.env.NULLSPEND_API_KEY!,
  },
});

// Use exactly as before — no other code changes
const message = await anthropic.messages.create({
  model: "claude-sonnet-4-20250514",
  max_tokens: 1024,
  messages: [{ role: "user", content: "Hello" }],
});

Python

from anthropic import Anthropic
import os

client = Anthropic(
    base_url="https://proxy.nullspend.dev",
    default_headers={
        "X-NullSpend-Key": os.environ["NULLSPEND_API_KEY"],
    },
)

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)

cURL

curl https://proxy.nullspend.dev/v1/messages \
  -H "x-api-key: $ANTHROPIC_API_KEY" \
  -H "X-NullSpend-Key: $NULLSPEND_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Tip: Use the full model name including the date suffix (e.g., claude-sonnet-4-20250514, not just claude-sonnet-4). The proxy resolves model aliases, but the full name ensures accurate pricing.

Step 3: Check the Dashboard

Open the NullSpend dashboard. Cost events appear within seconds of each request completing. You'll see:

Daily spend chart — cost over time
Model breakdown — which models cost the most
Per-key breakdown — costs attributed to each API key

Cache Token Tracking

NullSpend tracks Anthropic cache tokens separately. If you use prompt caching, the dashboard breaks down cost into:

Input tokens — standard input pricing
Cache read tokens — discounted (typically 90% cheaper than input)
Cache creation tokens — premium (typically 25% more than input)

This gives you accurate cost attribution even with heavy caching.

What's Next

Set a budget — Go to Budgets → Create Budget. The proxy blocks requests with 429 when the ceiling is hit.
Add tags — Attribute costs to teams or features with the X-NullSpend-Tags header.
Configure webhooks — Get notified on cost events, budget thresholds, and velocity alerts.
OpenAI too? — OpenAI Quickstart
Claude Agent SDK? — Claude Code Quickstart

Troubleshooting

401 Unauthorized Your X-NullSpend-Key header is missing, malformed, or the key has been revoked. Verify the key in Settings.

429 Too Many Requests Either a budget ceiling was hit (check the error.code field — budget_exceeded, velocity_exceeded, or rate_limited) or you've exceeded the rate limit (600 req/min per key). See the error reference.

Costs don't appear in the dashboard Cost logging is asynchronous. Wait a few seconds and refresh. If costs still don't appear, verify the request completed successfully in your application logs.

Streaming doesn't work The proxy transparently passes through SSE streams. Anthropic uses named-event SSE (event: message_start, event: content_block_delta, etc.) — the proxy handles this natively. Cost is calculated from the final message_delta event's usage data.

Anthropic Quickstart