# This file defines per-token pricing for AI models available in GitHub Copilot.
# Prices are per 1 million tokens.
#
# Please keep this list sorted by provider, then alphabetically by model name.
# Providers in this order: OpenAI, Anthropic, Google, xAI, Fine-tuned (GitHub).
#
# Column keys:
# - model: The model name.
# - provider: The model provider (openai, anthropic, google, xai, github).
# - release_status: "GA" or "Public preview".
# - category: The model category or task area.
# - input: Input token price per 1M tokens.
# - cached_input: Cached input token price per 1M tokens.
# - output: Output token price per 1M tokens.
# - cache_write: Cache write price per 1M tokens (Anthropic only).
# - notes: Optional notes about the model.

# OpenAI
- model: 'GPT-4.1[^1]'
  provider: openai
  release_status: GA
  category: Versatile
  input: $2.00
  cached_input: $0.50
  output: $8.00

- model: 'GPT-5 mini[^1]'
  provider: openai
  release_status: GA
  category: Lightweight
  input: $0.25
  cached_input: $0.025
  output: $2.00

- model: GPT-5.2
  provider: openai
  release_status: GA
  category: Versatile
  input: $1.75
  cached_input: $0.175
  output: $14.00

- model: GPT-5.2-Codex
  provider: openai
  release_status: GA
  category: Powerful
  input: $1.75
  cached_input: $0.175
  output: $14.00

- model: GPT-5.3-Codex
  provider: openai
  release_status: GA
  category: Powerful
  input: $1.75
  cached_input: $0.175
  output: $14.00

- model: 'GPT-5.4[^2]'
  provider: openai
  release_status: GA
  category: Versatile
  input: $2.50
  cached_input: $0.25
  output: $15.00
  notes: "Prompts \u2264272K tokens"

- model: GPT-5.4 mini
  provider: openai
  release_status: GA
  category: Lightweight
  input: $0.75
  cached_input: $0.075
  output: $4.50

- model: GPT-5.4 nano
  provider: openai
  release_status: GA
  category: Lightweight
  input: $0.20
  cached_input: $0.02
  output: $1.25

- model: GPT-5.5
  provider: openai
  release_status: GA
  category: Powerful
  input: $5.00
  cached_input: $0.50
  output: $30.00

# Anthropic
- model: Claude Haiku 4.5
  provider: anthropic
  release_status: GA
  category: Versatile
  input: $1.00
  cached_input: $0.10
  output: $5.00
  cache_write: $1.25

- model: Claude Sonnet 4
  provider: anthropic
  release_status: GA
  category: Versatile
  input: $3.00
  cached_input: $0.30
  output: $15.00
  cache_write: $3.75

- model: Claude Sonnet 4.5
  provider: anthropic
  release_status: GA
  category: Versatile
  input: $3.00
  cached_input: $0.30
  output: $15.00
  cache_write: $3.75

- model: Claude Sonnet 4.6
  provider: anthropic
  release_status: GA
  category: Versatile
  input: $3.00
  cached_input: $0.30
  output: $15.00
  cache_write: $3.75

- model: Claude Opus 4.5
  provider: anthropic
  release_status: GA
  category: Powerful
  input: $5.00
  cached_input: $0.50
  output: $25.00
  cache_write: $6.25

- model: Claude Opus 4.6
  provider: anthropic
  release_status: GA
  category: Powerful
  input: $5.00
  cached_input: $0.50
  output: $25.00
  cache_write: $6.25

- model: Claude Opus 4.7
  provider: anthropic
  release_status: GA
  category: Powerful
  input: $5.00
  cached_input: $0.50
  output: $25.00
  cache_write: $6.25

# Google
- model: 'Gemini 2.5 Pro[^5]'
  provider: google
  release_status: GA
  category: Powerful
  input: $1.25
  cached_input: $0.125
  output: $10.00
  notes: "Prompts \u2264200K tokens"

- model: 'Gemini 3 Flash[^6]'
  provider: google
  release_status: Public preview
  category: Lightweight
  input: $0.50
  cached_input: $0.05
  output: $3.00
  notes: No long-context surcharge

- model: 'Gemini 3.1 Pro[^5]'
  provider: google
  release_status: Public preview
  category: Powerful
  input: $2.00
  cached_input: $0.20
  output: $12.00
  notes: "Prompts \u2264200K tokens"

- model: 'Gemini 3.5 Flash'
  provider: google
  release_status: GA
  category: Lightweight
  input: $1.50
  cached_input: $0.15
  output: $9.00

# Fine-tuned (GitHub)
- model: 'Raptor mini[^7]'
  provider: github
  release_status: Public preview
  category: Versatile
  input: $0.25
  cached_input: $0.025
  output: $2.00
  notes: Uses GPT-5 mini pricing

- model: 'Goldeneye[^8]'
  provider: github
  release_status: Public preview
  category: Powerful
  input: $1.25
  cached_input: $0.125
  output: $10.00
  notes: Uses GPT-5.1-Codex pricing