GitHub Agentic Workflows

Model Aliases & Multipliers

This page lists the built-in model aliases and the per-model Effective Token (ET) multipliers used by GitHub Agentic Workflows.

[!CAUTION] The multiplier values shown on this page are approximations. They are used solely for the purpose of normalizing token usage across models into a single comparable metric (Effective Tokens) and do not represent precise cost ratios. Values may be inaccurate for specific model versions and may become out of date as providers update their offerings. Do not use these numbers for billing or financial calculations.

Model Aliases

Section titled “Model Aliases”

Model aliases let you write engine: copilot with a human-friendly model name such as sonnet or mini, and gh-aw resolves it to the best available concrete model at compile time. Each alias holds an ordered list of patterns; the first pattern that matches an available model wins.

For details on the alias syntax, fallback resolution algorithm, and how to define your own aliases in workflow frontmatter, see the Model Alias Format Specification.

Vendor Aliases

Section titled “Vendor Aliases”

Vendor aliases map a short name to one or more provider-scoped glob patterns. The Copilot gateway is always tried first.

AliasFallback patterns (tried in order)
sonnetcopilot/*sonnet*, anthropic/*sonnet*
sonnet-6xcopilot/*sonnet-4.5*, copilot/*sonnet-4.6*, copilot/*sonnet-4-5-*, anthropic/*sonnet-4-5-*, copilot/*sonnet-4-6*, anthropic/*sonnet-4-6*
haikucopilot/*haiku*, anthropic/*haiku*
opuscopilot/*opus*, anthropic/*opus*
gpt-5copilot/gpt-5*, openai/gpt-5*
gpt-5.5copilot/gpt-5.5*, openai/gpt-5.5*
gpt-5.4copilot/gpt-5.4*, openai/gpt-5.4*
gpt-5.3copilot/gpt-5.3*, openai/gpt-5.3*
gpt-5.2copilot/gpt-5.2*, openai/gpt-5.2*
gpt-5-minicopilot/gpt-5*mini*, openai/gpt-5*mini*
gpt-5-nanocopilot/gpt-5*nano*, openai/gpt-5*nano*
gpt-5-codexcopilot/gpt-5*codex*, openai/gpt-5*codex*
codingcopilot/gpt-5*codex*, openai/gpt-5*codex*, gpt-5-codex
mai-codecopilot/MAI-Code*, copilot/mai-code*, openai/MAI-Code*
gpt-5-procopilot/gpt-5*pro*, openai/gpt-5*pro*
reasoningcopilot/o1*, copilot/o3*, copilot/o4*, openai/o1*, openai/o3*, openai/o4*
gemini-flashcopilot/gemini-*flash*, google/gemini-*flash*, gemini/gemini-*flash*
gemini-flash-litecopilot/gemini-*flash*lite*, google/gemini-*flash*lite*, gemini/gemini-*flash*lite*
gemini-procopilot/gemini-*pro*, google/gemini-*pro*, gemini/gemini-*pro*
visioncopilot/gemini-*image*, gemini/gemini-*image*, copilot/gemini-*flash*, gemini/gemini-*flash*
gemmacopilot/gemma*, google/gemma*, gemini/gemma*
deep-researchcopilot/deep-research*, copilot/o3-deep-research*, copilot/o4-mini-deep-research*, google/deep-research*, gemini/deep-research*, openai/o3-deep-research*, openai/o4-mini-deep-research*
anycopilot/*, anthropic/*, openai/*, google/*, gemini/*
gemini-3-procopilot/gemini-3*pro*, google/gemini-3*pro*, google/nano-banana*, gemini/gemini-3*pro*
gemini-3-flashcopilot/gemini-3*flash*, google/gemini-3*flash*, gemini/gemini-3*flash*
gemini-3.1-procopilot/gemini-3.1*pro*, google/gemini-3.1*pro*, gemini/gemini-3.1*pro*
gemini-3.1-flashcopilot/gemini-3.1*flash*, google/gemini-3.1*flash*, gemini/gemini-3.1*flash*
gemini-3.5-flashcopilot/gemini-3.5*flash*, google/gemini-3.5*flash*, gemini/gemini-3.5*flash*
antigravitycopilot/antigravity*, google/antigravity*, gemini/antigravity*
nano-bananacopilot/nano-banana*, google/nano-banana*, gemini/nano-banana*
computer-usecopilot/*computer-use*, google/*computer-use*, gemini/*computer-use*, openai/*computer-use*
roboticscopilot/*robotics*, google/*robotics*, gemini/*robotics*

Meta-Aliases

Section titled “Meta-Aliases”

Meta-aliases reference other aliases by name. They are resolved recursively until a concrete pattern is reached.

Meta-aliasExpands to
opusplanopus?effort=high
smallmini
minihaiku → gpt-5-mini → gpt-5-nano → gemini-flash-lite
largesonnet → gpt-5-pro → gpt-5 → gemini-pro
agentsonnet-6x → gpt-5.4 → gpt-5.3 → gemini-pro → any
small-agenthaiku → gpt-5-mini → gemini-flash → any
copilotagent
claudeagent
codexagent
geminiagent
summarizationhaiku → gpt-5-mini → gemini-flash-lite → mini

Model Multipliers

Section titled “Model Multipliers”

Effective Token multipliers scale the weighted token total for each model relative to the reference model (claude-sonnet-4.5, multiplier = 1.0). A multiplier of 5.0 means that a run on that model counts as five times as many Effective Tokens as the same run on the reference model.

See the Effective Tokens Specification for the full formula.

Token Class Weights

Section titled “Token Class Weights”

Before per-model multipliers are applied, raw token counts are weighted by token class:

Token classDefault weight
Input1
Cached Input0.1
Output4
Reasoning4
Cache Write1

Per-Model Multipliers

Section titled “Per-Model Multipliers”

Anthropic

Section titled “Anthropic”
ModelMultiplier
claude-haiku-4-50.33
claude-haiku-4.50.33
claude-haiku-4-5-202510010.33
claude-3-5-haiku0.1
claude-3-haiku0.1
claude-sonnet-41
claude-sonnet-4-202505141
claude-sonnet-4-56
claude-sonnet-4.56
claude-sonnet-4-5-202509296
claude-sonnet-4-69
claude-sonnet-4.69
claude-3-5-sonnet1
claude-3-7-sonnet1
claude-3-sonnet1
claude-opus-45
claude-opus-4-202505145
claude-opus-4-15
claude-opus-4-1-202508055
claude-opus-4-515
claude-opus-4-5-2025110115
claude-opus-4-627
claude-opus-4-727
claude-opus-4-827
claude-opus-4.515
claude-opus-4.627
claude-opus-4.6-fast27
claude-opus-4.727
claude-opus-4.827
claude-3-5-opus5
claude-3-opus5

OpenAI

Section titled “OpenAI”
ModelMultiplier
gpt-4o-2024-05-130.33
gpt-4o-2024-08-060.33
gpt-4o-2024-11-200.33
gpt-4o-mini-2024-07-180.33
gpt-4.1-2025-04-141
gpt-41-copilot1
gpt-4.1-mini1
gpt-4.1-nano1
gpt-4-turbo1
gpt-41
gpt-4-06131
gpt-4-o-preview1
gpt-3.5-turbo0
gpt-3.5-turbo-06130
gpt-51
gpt-5-2025-08-071
gpt-5-search-api1
gpt-5-search-api-2025-10-141
gpt-5-chat-latest1
gpt-5-mini0.33
gpt-5-mini-2025-08-070.33
gpt-5-nano0.05
gpt-5-nano-2025-08-070.05
gpt-5-pro2
gpt-5-pro-2025-10-062
gpt-5.13
gpt-5.1-2025-11-133
gpt-5.1-chat-latest3
gpt-5-codex1
gpt-5.1-codex3
gpt-5.1-codex-mini0.33
gpt-5.1-codex-max3
gpt-5.1-codex-max-customsummarizer3
gpt-5.23
gpt-5.2-2025-12-113
gpt-5.2-chat-latest3
gpt-5.2-codex3
gpt-5.2-pro3
gpt-5.2-pro-2025-12-113
gpt-5.3-chat-latest3
gpt-5.3-codex6
gpt-5.3-codex-api-preview6
gpt-5.3-codex-api-preview-preambles6
gpt-5.46
gpt-5.4-2026-03-056
gpt-5.4-mini6
gpt-5.4-mini-2026-03-176
gpt-5.4-nano-2026-03-176
gpt-5.4-pro6
gpt-5.4-pro-2026-03-056
gpt-5.557
gpt-5.5-2026-04-2357
gpt-5.5-pro2
gpt-5.5-pro-2026-04-232

OpenAI Reasoning

Section titled “OpenAI Reasoning”
ModelMultiplier
o13
o1-2024-12-173
o1-mini0.5
o1-pro10
o1-pro-2025-03-1910
o33
o3-2025-04-163
o3-mini0.5
o3-mini-2025-01-310.5
o3-pro10
o3-pro-2025-06-1010
o3-deep-research3
o3-deep-research-2025-06-263
o4-mini0.5
o4-mini-2025-04-160.5
o4-mini-deep-research0.5
o4-mini-deep-research-2025-06-260.5

Google

Section titled “Google”
ModelMultiplier
gemini-2.5-pro1
gemini-2.5-pro-preview-tts1
gemini-2.5-flash0.2
gemini-2.5-flash-native-audio-latest0.2
gemini-2.5-flash-native-audio-preview-09-20250.2
gemini-2.5-flash-native-audio-preview-12-20250.2
gemini-2.5-flash-preview-tts0.2
gemini-2.5-flash-image0.2
gemini-2.5-flash-lite0.1
gemini-2.0-flash0.1
gemini-2.0-flash-0010.1
gemini-2.0-flash-lite0.1
gemini-2.0-flash-lite-0010.1
gemini-1.5-pro1
gemini-1.5-flash0.1
gemini-flash-latest0.2
gemini-flash-lite-latest0.1
gemini-pro-latest1
gemini-3-flash-preview0.33
gemini-3-pro-preview6
gemini-3-pro-image6
gemini-3-pro-image-preview6
gemini-3.1-pro-preview6
gemini-3.1-pro-preview-customtools6
gemini-3.1-flash-live-preview0.1
gemini-3.1-flash-lite0.1
gemini-3.1-flash-lite-preview0.1
gemini-3.1-flash-image0.33
gemini-3.1-flash-image-preview0.33
gemini-3.1-flash-tts-preview0.1
gemini-3.5-flash14
gemini-2.5-computer-use-preview0.2
gemini-2.5-computer-use-preview-10-20250.2
gemini-robotics-er-1.5-preview0.2
gemini-robotics-er-1.6-preview0.2

Other

Section titled “Other”
ModelMultiplier
antigravity-preview-05-20261
nano-banana-pro-preview0.2
deep-research-max-preview-04-20261
deep-research-preview-04-20261
deep-research-pro-preview-12-20251
MAI-Code-1-Flash0.33
gemma-4-26b-a4b-it0.1
gemma-4-31b-it0.2
grok-code-fast-10.33
raptor-mini0.33
Community Feedback
llms.txt llms-full.txt Create Workflow Install Reference

Made with by GitHub Next & Microsoft Research Terms Privacy Security