specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: osmAPI
providerId: osmapi
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: true
tags:
  - AI
  - Anthropic
  - Gateway
  - LLM
  - OpenAI
  - Routing
description: osmAPI is a pay-as-you-go AI gateway that proxies requests to multiple LLM
  providers. Pricing is pass-through - cost is calculated based on the underlying provider
  (OpenAI, Anthropic, Google, Groq) per token and surfaced in each API response. There are
  free models with reduced rate limits and paid models accessed via prepaid credits. Credits
  are purchased in INR via Razorpay and converted to USD for usage deduction.
sources:
  - https://docs.osmapi.com/features/cost-breakdown
  - https://docs.osmapi.com/resources/rate-limits
  - https://www.osmapi.com/
plans:
  - id: osmapi-free-models
    name: Free Models
    type: freemium
    description: Free model access available to all users with capped throughput. Suitable
      for evaluation and development.
    entries:
      - label: Rate Limit
        name: rate_limit
        type: quota
        metric: request
        limit: 200
        timeFrame: minute
        geo: global
        unit: 1
        price: '$0'
        userMultiplied: false
      - label: Token Cost
        name: token_cost
        type: flat
        metric: token
        limit: -1
        timeFrame: usage
        geo: global
        unit: 1
        price: '$0'
        userMultiplied: false
    elements:
      - name: Access to Free Models
      - name: Smart Routing
      - name: Streaming
      - name: OpenAI-Compatible API
      - name: Anthropic Messages API
  - id: osmapi-paid-models
    name: Paid Models (Pay-As-You-Go)
    type: usage-based
    description: Pay-as-you-go access to paid models on OpenAI, Anthropic, Google, Groq, and
      14+ providers. Per-token cost is pass-through to the underlying provider's published
      rates and returned in each API response. Credits are prepaid via Razorpay.
    entries:
      - label: Per-Token Cost
        name: token_cost
        type: metered
        metric: token
        limit: -1
        timeFrame: usage
        geo: global
        unit: 1
        price: see provider per-model rates returned in API response
        userMultiplied: false
      - label: Rate Limit (default)
        name: rate_limit
        type: quota
        metric: request
        limit: 1000
        timeFrame: minute
        geo: global
        unit: 1
        price: included
        userMultiplied: false
      - label: Credit Top-Up
        name: credit_topup
        type: flat
        metric: INR
        limit: -1
        timeFrame: usage
        geo: global
        unit: 1
        price: prepaid in INR via Razorpay, converted to USD on use
        userMultiplied: false
    elements:
      - name: Access to All Paid Models
      - name: Cost in API Response
      - name: Smart Routing
      - name: Streaming
      - name: Function Calling
      - name: Web Search
      - name: Response Healing
      - name: Embeddings
      - name: Audio
      - name: Realtime Endpoints
  - id: osmapi-enterprise
    name: Enterprise
    type: enterprise
    description: Custom-negotiated tier with higher rate limits beyond the default 1000 rpm
      per key. Throughput depends on credit balance and upstream provider constraints.
    entries:
      - label: Rate Limit
        name: rate_limit
        type: quota
        metric: request
        limit: -1
        timeFrame: minute
        geo: global
        unit: 1
        price: contact sales
        userMultiplied: false
      - label: Per-Token Cost
        name: token_cost
        type: metered
        metric: token
        limit: -1
        timeFrame: usage
        geo: global
        unit: 1
        price: contact sales
        userMultiplied: false
    elements:
      - name: Negotiated Rate Limits
      - name: Volume Pricing
      - name: Priority Routing
maintainers:
  - FN: API Evangelist
    email: info@apievangelist.com
x-canonical:
  generated: '2026-10-08'
  vocabulary: all/0-working/economics/plans-canonical.yml
  models:
  - metered
  - subscription
  units:
  - request
  - token
  plans:
  - name: Free Models
    access: freemium
    model: subscription
    model_how: derived
    units:
    - request
    - token
    periods:
    - minute
    - per-use
  - name: Paid Models (Pay-As-You-Go)
    access: other
    model: metered
    model_how: stated
    units:
    - other
    - request
    - token
    periods:
    - minute
    - per-use
  - name: Enterprise
    access: enterprise
    model: metered
    model_how: derived
    units:
    - request
    - token
    periods:
    - minute
    - per-use