specification: API Commons Plans
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/Plans
provider: OpenAI
providerId: openai
created: '2026-05-04'
modified: '2026-08-27'
generated: '2026-08-27'
method: searched
source: https://developers.openai.com/api/docs/pricing
sources:
- https://developers.openai.com/api/docs/pricing
- https://openai.com/api/pricing/
- https://developers.openai.com/changelog/
- https://developers.openai.com/api/docs/guides/rate-limits
plan_count: 13
reconciled: true
supersedes: >-
  The 2026-05-04 bulk-sweep version of this file, which was written without fetching the pricing
  page and topped out at GPT-5.5 — three model generations behind by the time it was read. Every
  price below was read from https://developers.openai.com/api/docs/pricing on 2026-08-27 and cross-checked
  against the dated pricing entries in the changelog (changelog/openai-changelog.yml, 2026-07-30
  and 2026-08-21).
tags:
- AI
- LLM
- GPT
- Foundation Models
- Usage-Based Pricing
description: >-
  OpenAI sells API access purely on metered token consumption — there are no seat plans, no monthly
  subscription tiers and no free-forever allowance on the API itself. The plan axis is the MODEL
  and the discount axis is the SERVICE TIER, and the two multiply: the same model costs half as much
  through the Batch API and twice as much in Fast mode. Prices moved down three times in the four
  weeks before this read (GPT-5.6 Luna -80%, Terra -20% on 2026-07-30; Sol to $4/$20 on 2026-08-21),
  so any cached copy of this page ages badly. Note the one uplift: regional processing adds 10% for
  models released on or after 2026-03-05. Access tiers (Free through Tier 5) are a RATE-LIMIT construct,
  not a price construct, and live in rate-limits/openai-rate-limits.yml.
currency: USD
unit: per 1,000,000 tokens
service_tiers:
- name: Standard
  multiplier: 1.0
  note: Base pricing, synchronous.
- name: Batch
  multiplier: 0.5
  note: 50% discount on input and output. Asynchronous; cancellable in flight.
- name: Flex
  multiplier: 0.5
  note: Documented as similar to Batch pricing.
- name: Fast mode
  multiplier: 2.0
  note: 2x standard. Replaced Priority Processing on 2026-07-30; supports >272K token context.
- name: Ultrafast
  multiplier: null
  note: Announced 2026-08-13 for GPT-5.6 Sol, up to 14x faster, limited preview to select customers.
    No public price.
uplifts:
- name: Regional processing
  pct: 10
  note: >-
    10% uplift for models released on or after 2026-03-05. Selectable per request via prefixed domains
    for keys in projects with Global geography (changelog 2026-08-21).
enterprise:
  contact: https://openai.com/contact-sales/
  note: Custom rate limits above Tier 5 and enterprise agreements are arranged through sales; no
    public price.
plans:
- id: openai-gpt-5-6-sol
  name: GPT-5.6 Sol
  type: usage-based
  description: Flagship reasoning model. Price reduced 2026-08-21 and held through at least 2026-11-21.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '4.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '20.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
  - name: Ultrafast mode (limited preview, up to 14x faster)
- id: openai-gpt-5-6-terra
  name: GPT-5.6 Terra
  type: usage-based
  description: General-purpose flagship. Price reduced 20% on 2026-07-30.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '2.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '12.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
- id: openai-gpt-5-6-luna
  name: GPT-5.6 Luna
  type: usage-based
  description: Low-cost model. Price reduced 80% on 2026-07-30.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '0.20'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '1.20'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
- id: openai-gpt-5-5
  name: GPT-5.5
  type: usage-based
  description: Previous flagship general-purpose model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '5.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '30.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
- id: openai-gpt-5-5-pro
  name: GPT-5.5 Pro
  type: usage-based
  description: Highest reasoning tier.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '30.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '180.00'
    userMultiplied: false
  elements:
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
- id: openai-gpt-5
  name: GPT-5
  type: usage-based
  description: Prior generation flagship.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '1.25'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '10.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
- id: openai-gpt-5-mini
  name: GPT-5 Mini
  type: usage-based
  description: Prior generation small model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '0.25'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '2.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
  - name: Flex service tier
  - name: Fast mode (2x standard)
- id: openai-o3
  name: o3
  type: usage-based
  description: Reasoning model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '2.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '8.00'
    userMultiplied: false
  elements:
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
- id: openai-o3-mini
  name: o3 Mini
  type: usage-based
  description: Small reasoning model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '1.10'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '4.40'
    userMultiplied: false
  elements:
  - name: Tool use
  - name: Structured outputs
  - name: Batch API (50% discount)
- id: openai-o1
  name: o1
  type: usage-based
  description: Legacy reasoning model. Shutdown announced 2026-04-22 for 2026-10-23.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '15.00'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '60.00'
    userMultiplied: false
  elements:
  - name: Tool use
  - name: Batch API (50% discount)
  - name: DEPRECATED
- id: openai-gpt-4o
  name: GPT-4o
  type: usage-based
  description: Legacy multimodal model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '2.50'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '10.00'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Batch API (50% discount)
- id: openai-gpt-4o-mini
  name: GPT-4o Mini
  type: usage-based
  description: Legacy small multimodal model.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '0.15'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '0.60'
    userMultiplied: false
  elements:
  - name: Vision
  - name: Tool use
  - name: Batch API (50% discount)
- id: openai-gpt-3-5-turbo
  name: GPT-3.5 Turbo
  type: usage-based
  description: Legacy model. Shutdown announced 2026-04-22 for 2026-10-23.
  entries:
  - label: Input tokens
    name: input
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '0.50'
    userMultiplied: false
  - label: Output tokens
    name: output
    type: metered
    metric: tokens
    limit: -1
    timeFrame: usage
    geo: global
    unit: 1000000
    price: '1.50'
    userMultiplied: false
  elements:
  - name: Tool use
  - name: DEPRECATED
caveat: >-
  Only text input/output token prices are recorded per model. OpenAI also publishes separate rates
  for cached input, audio tokens, image tokens, embeddings, fine-tuning training, and the built-in
  tools (web search, file search, code interpreter). Those are not reproduced here rather than partially
  transcribed; the pricing page is the source of truth and is linked as `source`.