models-and-pricing.md

title	Models and pricing for {% data variables.product.prodname_copilot %}
intro	See per-token pricing for the models available in {% data variables.product.prodname_copilot %} and reference rates for additional usage across plans.
shortTitle	Models and pricing
versions	feature copilot
category	Learn about Copilot
allowTitleToDifferFromFilename	true
contentType	reference

How model pricing works

When you use {% data variables.product.prodname_copilot_short %}, the interaction consumes tokens: input tokens (what's sent to the model), output tokens (what the model generates), and cached tokens (context the model reuses or stores). Each token is priced based on the model used, and the total is converted into {% data variables.product.prodname_ai_credits_short %}, where 1 {% data variables.product.prodname_ai_credit_singular %} = {% data variables.product.prodname_ai_credits_value %}.

The cost of an interaction depends on two things: the model and the number of tokens consumed.

How {% data variables.product.prodname_copilot_short %} usage is tracked and billed depends on your plan type:

• Individual plans ({% data variables.copilot.copilot_free_short %}, {% data variables.copilot.copilot_pro_short %}, {% data variables.copilot.copilot_pro_plus_short %}, and {% data variables.copilot.copilot_max_short %}) include {% data variables.product.prodname_ai_credits %} allowances that vary by plan. For details, see AUTOTITLE.
• {% data variables.copilot.copilot_business_short %} and {% data variables.copilot.copilot_enterprise_short %} include per-user {% data variables.product.prodname_ai_credits %} allowances that are pooled at the billing entity level. For details, see AUTOTITLE.

When usage exceeds the included allowances for any {% data variables.product.prodname_copilot_short %} plan, additional usage is billed in {% data variables.product.prodname_ai_credits %} at the per-token rates shown in the pricing tables below (1 {% data variables.product.prodname_ai_credit_singular %} = {% data variables.product.prodname_ai_credits_value %}).

{% data reusables.copilot.ai-credits-mobile %}

Pricing tables

All prices are per 1 million tokens.

OpenAI

{% data reusables.copilot.extended-context-pricing %}

Model	Release status	Category	Tier	Threshold (input tokens)	Input	Cached input	Output
{% for entry in tables.copilot.models-and-pricing %}{% if entry.provider == "openai" %}
{{ entry.model }}	{{ entry.release_status }}	{{ entry.category }}	{{ entry.tier }}	{{ entry.threshold }}	{{ entry.input }}	{{ entry.cached_input }}	{{ entry.output }}
{% endif %}{% endfor %}

Anthropic

Anthropic models include a cache write cost in addition to cached input.

{% data reusables.copilot.model-fable-disabled %}

Model	Release status	Category	Input	Cached input	Cache write	Output
{% for entry in tables.copilot.models-and-pricing %}{% if entry.provider == "anthropic" %}
{{ entry.model }}	{{ entry.release_status }}	{{ entry.category }}	{{ entry.input }}	{{ entry.cached_input }}	{{ entry.cache_write }}	{{ entry.output }}
{% endif %}{% endfor %}

Google

{% data reusables.copilot.extended-context-pricing %}

Model	Release status	Category	Tier	Threshold (input tokens)	Input	Cached input	Output
{% for entry in tables.copilot.models-and-pricing %}{% if entry.provider == "google" %}
{{ entry.model }}	{{ entry.release_status }}	{{ entry.category }}	{{ entry.tier }}	{{ entry.threshold }}	{{ entry.input }}	{{ entry.cached_input }}	{{ entry.output }}
{% endif %}{% endfor %}

Fine-tuned ({% data variables.product.github %})

Model	Release status	Category	Input	Cached input	Output
{% for entry in tables.copilot.models-and-pricing %}{% if entry.provider == "github" %}
{{ entry.model }}	{{ entry.release_status }}	{{ entry.category }}	{{ entry.input }}	{{ entry.cached_input }}	{{ entry.output }}
{% endif %}{% endfor %}

Microsoft

Model	Release status	Category	Input	Cached input	Output
{% for entry in tables.copilot.models-and-pricing %}{% if entry.provider == "microsoft" %}
{{ entry.model }}	{{ entry.release_status }}	{{ entry.category }}	{{ entry.input }}	{{ entry.cached_input }}	{{ entry.output }}
{% endif %}{% endfor %}

Code completions

{% data reusables.copilot.tbb-completions %}

Pricing and usage cost considerations for {% data variables.copilot.copilot_code-review_short %}

For most {% data variables.product.prodname_copilot_short %} features, the model used for each interaction is visible to you, so you can reference the pricing tables above to estimate costs. {% data variables.copilot.copilot_code-review_short %} is an exception—the model is selected automatically and is not disclosed, so per-token costs may vary between reviews.

Each code review is billed in two ways: token consumption is billed in {% data variables.product.prodname_ai_credits_short %}, and the agentic infrastructure that powers the review consumes {% data variables.product.prodname_actions %} minutes.

{% data variables.product.prodname_actions %} minutes are attributed to the repository, and from there to the enterprise or cost center where applicable. {% data variables.product.prodname_ai_credits_short %} are charged to the person who requests the review, or to the author of a pull request where a policy automatically triggers a review. If neither has a {% data variables.product.prodname_copilot_short %} seat, usage is billed to the enterprise or cost center instead.

You can view your current {% data variables.product.prodname_actions %} usage for {% data variables.copilot.copilot_code-review_short %} in the following ways:

• {% data variables.product.prodname_actions %} metrics: Filter by the copilot-pull-request-reviewer workflow. See AUTOTITLE.
• Billing usage report: Filter by workflow_path using the value dynamic/agents/copilot-pull-request-reviewer. See AUTOTITLE.

Model multipliers for annual {% data variables.copilot.copilot_pro_short %} and {% data variables.copilot.copilot_pro_plus_short %} subscribers

{% data variables.copilot.copilot_pro_short %} and {% data variables.copilot.copilot_pro_plus_short %} subscribers on existing annual billing plans using the request-based billing model have different model multipliers. See AUTOTITLE.