Skip to calculator
ModelMeter
Version 1.0 · Verified pricing with context tiers

Know your AI bill before it arrives.

Estimate and compare AI API costs based on your actual usage.

Compare standard text API pricing across leading AI providers.

Standard text pricing, verified against official provider documentation.

Pricing verified from official provider documentation. Last review: Sep 18, 2026.

Usage Calculator

Enter your expected monthly volume. Every field accepts non-negative numbers only.

Scale

100K requests per month, moderate prompts.

Scenario

Start from the default estimate.

Preset values are illustrative examples only, not industry averages or benchmarks.

Total API requests per month

Prompt tokens sent per request

Some providers change pricing for long-context requests. ModelMeter applies supported pricing tiers automatically.

Completion tokens returned per request

Include billable reasoning / thinking tokens where your provider charges for them.

Usage Summary

Monthly Requests
100,000
Monthly Input Tokens
200M
Monthly Output Tokens
50M
Total Tokens
250M

Estimated Monthly Cost

Based on your usage and the verified pricing dataset.

Sorted by estimated monthly cost, lowest first.

Estimated Monthly Spend

Lowest
$100.00 / month
Highest
$2,250.00 / month
Potential cost difference
$2,150.00 / month

Based on your usage, the cheapest option costs $100.00 per month and the most expensive costs $2,250.00 per month.

Switching from the highest-cost option to the lowest-cost option could reduce the estimated monthly cost by $2,150.00 (95.6%).

Compared on estimated cost only. Model quality, latency and capability are not measured.

  • OpenAILowest estimated cost

    gpt-5.6-luna

    gpt-5.6-luna

    Estimated Monthly Cost

    $100.00

    Pricing tier
    Standard context
    Your average input
    2,000 tokens / request
    Pricing threshold
    272K tokens / request
    Input price
    $0.20 / 1M tokens
    Output price
    $1.20 / 1M tokens
    Estimated input cost
    $40.00
    Estimated output cost
    $60.00

    OpenAI bills requests above the 272K-token context boundary at this model's long-context rate. ModelMeter selects the tier from your average input tokens per request.

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

  • Google

    Gemini 3.1 Flash-Lite

    gemini-3.1-flash-lite

    Estimated Monthly Cost

    $125.00

    +$25.00 vs lowest-cost option · 25% higher than the lowest-cost option

    Pricing tier
    Standard context
    Input price
    $0.25 / 1M tokens
    Output price
    $1.50 / 1M tokens
    Estimated input cost
    $50.00
    Estimated output cost
    $75.00

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

  • Anthropic

    Claude Sonnet 5

    claude-sonnet-5

    Estimated Monthly Cost

    $900.00

    +$800.00 vs lowest-cost option · 800% higher than the lowest-cost option

    Pricing tier
    Standard context
    Input price
    $2.00 / 1M tokens
    Output price
    $10.00 / 1M tokens
    Estimated input cost
    $400.00
    Estimated output cost
    $500.00

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

  • OpenAI

    gpt-5.6-terra

    gpt-5.6-terra

    Estimated Monthly Cost

    $1,000.00

    +$900.00 vs lowest-cost option · 900% higher than the lowest-cost option

    Pricing tier
    Standard context
    Your average input
    2,000 tokens / request
    Pricing threshold
    272K tokens / request
    Input price
    $2.00 / 1M tokens
    Output price
    $12.00 / 1M tokens
    Estimated input cost
    $400.00
    Estimated output cost
    $600.00

    OpenAI bills requests above the 272K-token context boundary at this model's long-context rate. ModelMeter selects the tier from your average input tokens per request.

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

  • Google

    Gemini 3.1 Pro Preview

    gemini-3.1-pro-preview

    Estimated Monthly Cost

    $1,000.00

    +$900.00 vs lowest-cost option · 900% higher than the lowest-cost option

    Pricing tier
    Standard context
    Your average input
    2,000 tokens / request
    Pricing threshold
    200K tokens / request
    Input price
    $2.00 / 1M tokens
    Output price
    $12.00 / 1M tokens
    Estimated input cost
    $400.00
    Estimated output cost
    $600.00

    Google bills prompts above 200,000 tokens at this model's higher tier. ModelMeter selects the tier from your average input tokens per request.

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

  • Anthropic

    Claude Opus 5

    claude-opus-5

    Estimated Monthly Cost

    $2,250.00

    +$2,150.00 vs lowest-cost option · 2,150% higher than the lowest-cost option

    Pricing tier
    Standard context
    Input price
    $5.00 / 1M tokens
    Output price
    $25.00 / 1M tokens
    Estimated input cost
    $1,000.00
    Estimated output cost
    $1,250.00

    Pricing mode: Standard

    Pricing source: Official provider documentation

    Last checked: Sep 18, 2026

Pricing Sources

Every price in the dataset is read from the provider's own pricing documentation. Third-party price aggregators are not used.

Pricing data should be verified against official provider documentation.

Last pricing review: Sep 18, 2026

Known Limitations

About ModelMeter

ModelMeter turns expected token usage into a monthly cost estimate and compares that estimate across AI models. It is built as a small, focused developer tool: enter your volume, read the breakdown, and sanity-check a budget before the invoice arrives.

Pricing is maintained by hand from each provider's official documentation. Every entry records the model identifier, the standard-tier input and output prices, the pricing mode, the source URL and the date it was last verified — see Pricing Sources.

ModelMeter compares estimated cost only. It does not rank model quality.

The source code and the full pricing dataset are public in the ModelMeter GitHub repository (opens in a new tab).