Quickstart

Start measuring the carbon footprint of your LLM calls in under 5 minutes. The API is REST + JSON — no OpenAPI bundle yet, but paths and payloads are stable and documented below.

1. Create your account

2. Try /v1/estimate

The /v1/estimate endpoint does not require authentication and stores nothing.

curl -X POST https://carbon-llm.com/api/v1/estimate \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "prompt_tokens": 450,
    "completion_tokens": 120
  }'

3. Integrate /v1/track in production

Add these two lines after each LLM call to track emissions.

# After your LLM call
requests.post('https://carbon-llm.com/api/v1/track',
    headers={'Authorization': 'Bearer isv_live_xxxx'},
    json={'tenant_id': 'client-xyz', 'model': 'gpt-4o',
          'prompt_tokens': response.usage.prompt_tokens,
          'completion_tokens': response.usage.completion_tokens})

Note: Prompts never pass through our servers. Only metadata (model, token counts) is sent.

Back to home Try the API