token-dollars

Calculate costs for AI model API usage based on token consumption and pricing.

Last verified: June 2026 from openai.com/api/pricing & ai.google.dev/gemini-api/docs/pricing.
All prices are USD per 1M tokens unless noted otherwise.
Update this file when providers reprice — prices change frequently.

Installation

npm i token-dollars

Usage

Import

import { Cost, Store } from 'token-dollars';

Text model

const result = Cost('gpt-4.1', {
  total_text_input_tokens: 1000000,
  total_text_output_tokens: 500000,
  total_cached_text: 200000,
});

console.log(result);
// {
//   model: 'gpt-4.1',
//   total: 3.9,
//   breakdown: {
//     textInputCost: 1.6,   // (1M - 200K cached) × $2.00/M
//     cachedTextInputCost: 0.1,
//     textOutputCost: 4,
//     audioInputCost: 0,
//     cachedAudioInputCost: 0,
//     audioOutputCost: 0,
//     videoInputCost: 0,
//     imageInputCost: 0,
//     cachedImageInputCost: 0,
//     whisperCost: 0
//   }
// }

Realtime / audio model

const result = Cost('gpt-realtime-2', {
  total_text_input_tokens: 500000,
  total_audio_input_tokens: 1000000,
  total_audio_output_tokens: 500000,
  total_cached_audio: 100000,
  input_Transcript_Duration_whisper: 120, // 2 minutes
});

console.log(result.total);

Time-billed model (translate / whisper realtime)

These models bill per second, not per token. Pass duration_seconds instead of token counts:

const result = Cost('gpt-realtime-translate', {
  duration_seconds: 300, // 5 minutes
});

console.log(result.total); // 300/60 × $0.034 = $0.17
// { model: 'gpt-realtime-translate', total: 0.17, breakdown: { timeCost: 0.17 } }

Gemini Live API

const result = Cost('gemini-live-2.5-flash', {
  total_text_input_tokens: 200000,
  total_audio_input_tokens: 50000,
  total_audio_output_tokens: 30000,
  total_video_input_tokens: 10000,
});

Usage Object

All properties are optional and default to 0.

| Property | Description | |---|---| | total_text_input_tokens | Text input tokens | | total_cached_text | Cached text input tokens (cheaper rate) | | total_text_output_tokens | Text output tokens | | total_audio_input_tokens | Audio input tokens | | total_cached_audio | Cached audio input tokens | | total_audio_output_tokens | Audio output tokens | | total_video_input_tokens | Video input tokens (Gemini) | | total_image_input_tokens | Image input tokens | | total_image_cached_input_tokens | Cached image input tokens | | input_Transcript_Duration_whisper | Whisper transcription duration in seconds | | duration_seconds | Session duration for time-billed models |

Supported Models

OpenAI — GPT-5.x Flagship

| Model | Input | Cached | Output | |---|---|---|---| | gpt-5.5 | $5.00 | $0.50 | $30.00 | | gpt-5.5-pro | $30.00 | — | $180.00 | | gpt-5.4 | $2.50 | $0.25 | $15.00 | | gpt-5.4-mini | $0.75 | $0.075 | $4.50 | | gpt-5.4-nano | $0.20 | $0.02 | $1.25 | | gpt-5.4-pro | $30.00 | — | $180.00 |

OpenAI — GPT-5 (original)

| Model | Input | Cached | Output | |---|---|---|---| | gpt-5 | $1.25 | $0.125 | $10.00 | | gpt-5-mini | $0.25 | $0.025 | $2.00 | | gpt-5-nano | $0.05 | $0.005 | $0.40 | | gpt-5-pro | $15.00 | — | $120.00 | | gpt-5-chat-latest | $1.25 | $0.125 | $10.00 |

OpenAI — GPT-4.1

| Model | Input | Cached | Output | |---|---|---|---| | gpt-4.1 | $2.00 | $0.50 | $8.00 | | gpt-4.1-mini | $0.40 | $0.10 | $1.60 | | gpt-4.1-nano | $0.10 | $0.025 | $0.40 |

OpenAI — GPT-4o (legacy)

| Model | Input | Cached | Output | |---|---|---|---| | gpt-4o | $2.50 | $1.25 | $10.00 | | gpt-4o-2024-05-13 | $5.00 | — | $15.00 | | gpt-4o-mini | $0.15 | $0.075 | $0.60 |

OpenAI — o-series Reasoning

| Model | Input | Cached | Output | |---|---|---|---| | o3 | $2.00 | $0.50 | $8.00 | | o3-pro | $20.00 | — | $80.00 | | o4-mini | $1.10 | $0.275 | $4.40 | | o1 | $15.00 | $7.50 | $60.00 | | o1-mini | $3.00 | $1.50 | $12.00 | | o1-pro | $150.00 | — | $600.00 |

OpenAI — Realtime (token-billed)

| Model | Text In | Text Out | Audio In | Audio Out | |---|---|---|---|---| | gpt-realtime-2 | $4.00 | $24.00 | $32.00 | $64.00 | | gpt-realtime-1.5 | $4.00 | $16.00 | $32.00 | $64.00 | | gpt-realtime-mini | $0.60 | $2.40 | $10.00 | $20.00 |

OpenAI — Realtime (time-billed) — pass duration_seconds

| Model | Rate | |---|---| | gpt-realtime-translate | $0.034/min | | gpt-realtime-whisper | $0.017/min |

OpenAI — Transcription

| Model | Input | Output | Est. cost | |---|---|---|---| | gpt-4o-transcribe | $2.50 | $10.00 | ~$0.006/min | | gpt-4o-mini-transcribe | $1.25 | $5.00 | ~$0.003/min | | whisper-1 | — | — | $0.006/min |

Google Gemini — Standard

| Model | Input | Cached | Output | |---|---|---|---| | gemini-3.5-flash | $1.50 | $0.15 | $9.00 | | gemini-3.1-pro | $2.00 | $0.20 | $12.00 | | gemini-3.1-flash-lite | $0.25 | $0.025 | $1.50 | | gemini-2.5-pro | $1.25 | $0.125 | $10.00 | | gemini-2.5-flash | $0.30 | $0.03 | $2.50 | | gemini-2.5-flash-lite | $0.10 | $0.01 | $0.40 |

Google Gemini — Live / Realtime API

| Model | Text In | Audio In | Audio Out | Video In | |---|---|---|---|---| | gemini-live-2.5-flash | $0.30 | $1.00 | $4.00 | $2.00 | | gemini-live-2.5-pro | $1.25 | $3.00 | $12.00 | $6.00 | | gemini-2.5-flash-native-audio-latest | $0.50 | $3.00 | $12.00 | $3.00 |

API

Cost(model, usage)

Calculates the total cost for a given model based on usage.

Returns:

{
  model: string,
  total: number,          // rounded to 6 decimal places
  breakdown: {
    textInputCost: number,
    cachedTextInputCost: number,
    textOutputCost: number,
    audioInputCost: number,
    cachedAudioInputCost: number,
    audioOutputCost: number,
    videoInputCost: number,
    imageInputCost: number,
    cachedImageInputCost: number,
    whisperCost: number,
    timeCost?: number,    // only for time-billed models
  }
}

Returns { model, total: 0, breakdown: {} } for unknown models — no throw.

Store

An object containing pricing for all supported models. Each entry includes:

• pricing — Rates per 1M tokens (or per minute for time-based models)
• asp — Provider: "openai" or "google"
• billingUnit — "time" for per-minute models, otherwise token-based
• wssUrl — WebSocket endpoint (realtime models only)

Pricing Notes

• All token rates are per 1M tokens in USD
• Cached tokens are billed at a reduced rate (typically 50–90% less than base input)
• Whisper/transcription costs are per minute of audio
• Time-billed realtime models use duration_seconds, not token counts
• Gemini Live API bills per turn on the full session context window
• Reasoning models (o-series, Gemini 2.5+) include internal thinking tokens in output costs
• Costs are rounded to 6 decimal places

License

ISC