// use case

AI / LLM Platform

"I sell AI model access. Billing is per token (input/output separately) and per image generation."

Token-based billing for AI platforms. Separate metering for input and output tokens with model-level filtering, plus per-image charges for generation endpoints.

// billable metrics

What you meter

metric

Input Tokens

code: input_tokens

aggregation: sum

field: input_token_count

metric

Output Tokens

code: output_tokens

aggregation: sum

field: output_token_count

metric

Image Generations

code: image_generations

aggregation: count

// features

Feature entitlements

Fine Tuning

code: fine_tuning

type: boolean

Model Access Tier

code: model_access_tier

type: custom

which models are available

Max Context Length

code: max_context_length

type: quantity

// plans

Pricing tiers

Free

$0/mo

ai_free · monthly

Charges

input_tokens → package

$0 per 1M tokens (capped via entitlement)

output_tokens → package

$0 per 1M tokens

Entitlements

✗ fine_tuning

= model_access_tier basic

= max_context_length 4096

Developer

$20/mo

ai_developer · monthly

Charges

input_tokens → graduated

0–1M @ $0.50/1M, 1M–10M @ $0.30/1M

output_tokens → graduated

0–1M @ $1.50/1M, 1M–10M @ $1.00/1M

image_generations → standard

$0.04 each

Entitlements

✗ fine_tuning

= model_access_tier standard

= max_context_length 16384

Scale

$200/mo

ai_scale · monthly

Charges

input_tokens → volume

0–10M @ $0.20/1M, 10M+ @ $0.10/1M

output_tokens → volume

0–10M @ $0.80/1M, 10M+ @ $0.50/1M

image_generations → standard

$0.02 each

Entitlements

✓ fine_tuning

= model_access_tier all

= max_context_length 128000

// add-ons

Optional add-ons

dedicated_endpoint

Dedicated model inference endpoint

$100

content_moderation

Automated content safety filtering

$25

Ready to implement ai / llm platform billing? bxb handles all of this out of the box.

> Services ← Home