Clone Mistral pricing

Replicate Mistral’s per-token pricing model with Lago.

In this article, you will learn how Mistral is using Lago to build a billing system based on AI tokens. This template is suitable for Large Language Model (LLM) and Generative AI companies whose pricing can vary based on the application or model used.

Pricing structure

For Mistral, pricing depends on the language model used. Here are several price points they offer: Prices are per 1M tokens used. You can think of tokens as pieces of words (learn more here).”

Models	Input	Output
mistral-small	$1 / 1M tokens	$3 / 1M tokens
mistral-medium	$2.7 / 1M tokens	$8.1 / 1M tokens
mistral-large	$4 / 1M tokens	$12 / 1M tokens

Get started

1. Set up Lago

   cURL

   LAGO_URL="https://api.getlago.com"
   API_KEY="__API_KEY__"
   

2. Create a billable metric

   Create a single billable metric to track token usage across different models and input/output types.

   1. Add a unique code for your billable metric
   2. Set aggregation_type to sum_agg
   3. Set field_name to tokens
   4. Set recurring to false
   5. Set filters to distinguish between models and token types

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/billable_metrics" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "billable_metric": {
         "name": "AI Tokens",
         "code": "__BILLABLE_METRIC_CODE__",
         "description": "Token usage across Mistral AI models",
         "aggregation_type": "sum_agg",
         "field_name": "tokens",
         "recurring": false,
         "filters": [
           {
             "key": "model",
             "values": ["mistral-small", "mistral-medium", "mistral-large"]
           },
           {
             "key": "type", 
             "values": ["input", "output"]
           }
         ]
       }
     }'
   

   Refer to the API reference and guide on filters to learn more.

3. Create a plan

   Create a plan to price packages of tokens used with Mistral’s specific pricing tiers.

   1. Add a unique code for your plan
   2. Under charges, configure pricing tiers for different models and token types

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/plans" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "plan": {
         "name": "Mistral AI Pricing",
         "code": "__PLAN_CODE__",
         "amount_cents": 0,
         "amount_currency": "USD",
         "interval": "monthly",
         "pay_in_advance": false,
         "charges": [
           {
             "billable_metric_id": "__BILLABLE_METRIC_ID__",
             "charge_model": "package",
             "filters": [
               {
                 "invoice_display_name": "mistral-small-input",
                 "properties": {
                   "amount": "1",
                   "free_units": 0,
                   "package_size": 1000000
                 },
                 "values": {
                   "model": ["mistral-small"],
                   "type": ["input"]
                 }
               },
               {
                 "invoice_display_name": "mistral-small-output", 
                 "properties": {
                   "amount": "3",
                   "free_units": 0,
                   "package_size": 1000000
                 },
                 "values": {
                   "model": ["mistral-small"],
                   "type": ["output"]
                 }
               },
               {
                 "invoice_display_name": "mistral-large-input",
                 "properties": {
                   "amount": "4", 
                   "free_units": 0,
                   "package_size": 1000000
                 },
                 "values": {
                   "model": ["mistral-large"],
                   "type": ["input"]
                 }
               },
               {
                 "invoice_display_name": "mistral-large-output",
                 "properties": {
                   "amount": "12",
                   "free_units": 0, 
                   "package_size": 1000000
                 },
                 "values": {
                   "model": ["mistral-large"],
                   "type": ["output"]
                 }
               } 
             ]
           }
         ]
       }
     }'
   

   Refer to the API reference and guide on charges with filters to learn more.

4. Create a customer

   Create a customer with a unique external_id.

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/customers" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "customer": {
         "external_id": "__EXTERNAL_CUSTOMER_ID__",
         "name": "Acme Inc",
         "email": "[e-mail]",
         "currency": "USD",
         "timezone": "America/New_York"
       }
     }'
   

   Refer to the API reference to create a customer.

5. Prepay usage with credits

   1. Create a wallet with the customer’s external_customer_id, set the wallet currency and rate_amount for credit conversion

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/wallets" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "wallet": {
         "name": "AI Credits",
         "external_customer_id": "__EXTERNAL_CUSTOMER_ID__",
         "currency": "USD",
         "paid_credits": "100.0",
         "granted_credits": "25.0",
         "rate_amount": "1.0"
       }
     }'
   

   Refer to the API reference to create a wallet.

   2. Create a subscription for the customer with the plan’s code.

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/subscriptions" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "subscription": {
         "external_customer_id": "__EXTERNAL_CUSTOMER_ID__",
         "plan_code": "__PLAN_CODE__",
         "external_id": "__EXTERNAL_SUBSCRIPTION_ID__"
       }
     }'
   

   Refer to the API reference to create a subscription.

6. Ingest usage via events

   Send usage events to Lago to track usage.

   1. Reference your billable metric with code
   2. Reference the customer’s subscription with external_subscription_id
   3. Include usage and filters data in properties

   cURL

   curl --location --request POST "$LAGO_URL/api/v1/events" \
     --header "Authorization: Bearer $API_KEY" \
     --header 'Content-Type: application/json' \
     --data-raw '{
       "event": {
         "transaction_id": "__TRANSACTION_ID__",
         "code": "__BILLABLE_METRIC_CODE__",
         "external_subscription_id": "__EXTERNAL_SUBSCRIPTION_ID__",
         "properties": {
           "tokens": 2000000,
           "model": "mistral-small",
           "type": "input"
         }
       }
     }'
   

   Refer to the API reference to create an event.

7. Monitor current usage

   Track real-time customer usage for the current billing period.

   cURL

   curl --location --request GET "$LAGO_URL/api/v1/customers/__EXTERNAL_CUSTOMER_ID__/current_usage?external_subscription_id=__EXTERNAL_SUBSCRIPTION_ID__" \
   --header 'Authorization: Bearer __API_KEY__' \
   --header 'Content-Type: application/json'
   

   Refer to the API reference to get the current usage.

Wrap-up

Per-token pricing offers flexibility and visibility, and allows LLM and Generative AI companies like Mistral to attract more customers. With Lago, you can create your own metric dimensions to adapt this template to your products and services.

Give it a try, click here to get started!