[model-inventory] Model alias inventory update - 2026-05-25

[github-actions]

Summary

Daily model inventory check completed on 2026-05-25. Most aliases and multipliers are up to date. Key findings include a missing claude-opus-4-7 multiplier key (dot vs dash ID variant), several dated GPT-4o/GPT-3.5 model IDs absent from multipliers, and two missing dedicated aliases (gpt-5.5 and gemini-3.5-flash) that warrant addition for consistency.

• Providers queried: OpenAI, Anthropic, Gemini, Copilot
• Total models found: 283 (Copilot: 31, OpenAI: 193, Anthropic: 9, Gemini: 50)
• Proposed alias changes: 2 new aliases
• Multiplier gaps found: 10 (mostly dated snapshot IDs + 1 variant key)

Provider Model Counts

Provider	Models Available	Status
openai	193	✅ ok
anthropic	9	✅ ok
gemini	50	✅ ok
copilot	31	✅ ok

Raw API Fields Discovered

Anthropic (raw.json):

• display_name, created_at, max_input_tokens, max_tokens
• capabilities: batch.supported, citations.supported, code_execution.supported, context_management.supported, effort.supported, image_input.supported, pdf_input.supported, structured_outputs.supported, thinking.supported

OpenAI (raw.json):

• Minimal fields: id, object, created (Unix timestamp), owned_by
• No token limits, billing, or capability flags in the models list endpoint

Gemini (raw.json):

• name, version, displayName, description
• inputTokenLimit, outputTokenLimit (useful as complexity proxy)
• supportedGenerationMethods (e.g. generateContent, batchGenerateContent)
• temperature, topP, topK, maxTemperature
• thinking boolean (extended reasoning flag — notable for Gemini 2.5 Flash, 3.x Pro)

Copilot (reflect endpoint):

• id, vendor, name
• capabilities.limits.max_context_window_tokens, capabilities.limits.max_output_tokens
• capabilities.supports.vision, capabilities.type
• billing.multiplier

Token Multiplier Analysis

Missing from model_multipliers.json

Model ID	Provider	Inferred Multiplier	Basis
claude-opus-4-7	copilot/anthropic	27.0	Docs table (Claude Opus 4.7); claude-opus-4.7 IS present with 27.0 — this is a dash vs dot ID variant
gpt-4o-2024-05-13	copilot	~0.33	Dated snapshot of gpt-4o (multiplier 0.33)
gpt-4o-2024-08-06	copilot	~0.33	Dated snapshot of gpt-4o
gpt-4o-2024-11-20	copilot	~0.33	Dated snapshot of gpt-4o
gpt-4o-mini-2024-07-18	copilot	~0.33	Dated snapshot of gpt-4o-mini
gpt-4-o-preview	copilot	~1.0	Legacy GPT-4 preview variant
gpt-4-0613	copilot	~1.0	Dated snapshot of gpt-4 (stored: 1.0)
gpt-3.5-turbo	copilot	~0	Legacy; not in docs billing table
gpt-3.5-turbo-0613	copilot	~0	Legacy snapshot
antigravity-preview-05-2026	gemini	unknown	New preview; no docs table entry

Note on embedding models: text-embedding-3-small, text-embedding-3-small-inference, text-embedding-ada-002 are listed in Copilot reflect but are embedding models, not generative — no ET multiplier applies.

Historical entries not currently returned

The following entries are in model_multipliers.json but were not returned by any live API. They should remain as historical records:

gpt-5-2025-08-07, gpt-5-search-api, gpt-5-chat-latest, gpt-5-mini-2025-08-07, gpt-5-nano-2025-08-07, gpt-5-pro-2025-10-06, gpt-5.1, gpt-5.1-codex, gpt-5.1-codex-mini, gpt-5.1-codex-max, gpt-5.2-pro, gpt-5.3-chat-latest, o1, o3, o3-mini, o3-deep-research, o4-mini, o4-mini-deep-research, o1-pro, o3-pro, claude-3-*, gemini-1.5-*, grok-code-fast-1, raptor-mini, and others.

Inferred vs stored discrepancies

All models present in both live APIs and model_multipliers.json match the GitHub docs billing table. No discrepancies found.

Notable: GPT-5.5 is listed in the docs table as New multiplier: TBD — the stored value of 7.5 is the current value and should be retained until the new value is published.

Proposed Alias Updates

1. Add gpt-5.5 alias

What: New dedicated alias gpt-5.5 matching gpt-5.5 family models.

Why: gpt-5.5 is now live in Copilot (multiplier 7.5, listed as "TBD" for new billing). Aliases gpt-5.4, gpt-5.3, gpt-5.2 exist for consistency; gpt-5.5 should follow the same pattern. Currently gpt-5 broadly matches it via copilot/gpt-5*, but an explicit alias allows workflows to pin to this generation.

Syntax:

{
  "gpt-5.5": ["copilot/gpt-5.5*", "openai/gpt-5.5*"]
}

2. Add gpt-5.2 alias

What: New dedicated alias gpt-5.2 matching gpt-5.2 family models.

Why: gpt-5.2 and gpt-5.2-codex are live in Copilot. The other gpt-5.x generation aliases (gpt-5.3, gpt-5.4, gpt-5.5) already exist; gpt-5.2 is missing for completeness.

Syntax:

{
  "gpt-5.2": ["copilot/gpt-5.2*", "openai/gpt-5.2*"]
}

3. Add gemini-3.5-flash alias

What: New dedicated alias for Gemini 3.5 Flash.

Why: gemini-3.5-flash is live in the Gemini API with a very high billing multiplier (14.0), making it distinct from other Flash models. The existing gemini-flash broad alias matches it, but given the multiplier difference from gemini-3-flash (0.33) and gemini-2.x-flash (0.1–0.2), a dedicated alias prevents accidental use by workflows using gemini-flash expecting low-cost flash models.

Syntax:

{
  "gemini-3.5-flash": ["copilot/gemini-3.5*flash*", "google/gemini-3.5*flash*", "gemini/gemini-3.5*flash*"]
}

Full Model Lists by Provider

Copilot (31 models)

claude-haiku-4.5, claude-opus-4.5, claude-opus-4.6, claude-opus-4.7,
claude-sonnet-4.5, claude-sonnet-4.6, gemini-2.5-pro, gpt-3.5-turbo,
gpt-3.5-turbo-0613, gpt-4, gpt-4-0613, gpt-4-o-preview, gpt-4.1,
gpt-4.1-2025-04-14, gpt-4o, gpt-4o-2024-05-13, gpt-4o-2024-08-06,
gpt-4o-2024-11-20, gpt-4o-mini, gpt-4o-mini-2024-07-18, gpt-5-mini,
gpt-5.2, gpt-5.2-codex, gpt-5.3-codex, gpt-5.4, gpt-5.4-mini,
gpt-5.4-nano, gpt-5.5, text-embedding-3-small,
text-embedding-3-small-inference, text-embedding-ada-002

Anthropic (9 models)

claude-haiku-4-5-20251001, claude-opus-4-1-20250805, claude-opus-4-20250514,
claude-opus-4-5-20251101, claude-opus-4-6, claude-opus-4-7,
claude-sonnet-4-20250514, claude-sonnet-4-5-20250929, claude-sonnet-4-6

OpenAI (193 models — top production models)

gpt-4o, gpt-4o-mini, gpt-4.1, gpt-4.1-2025-04-14, gpt-5.2, gpt-5.2-codex,
gpt-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.4-nano, gpt-5.5,
o1, o1-mini, o1-pro, o3, o3-mini, o3-pro, o3-deep-research,
o4-mini, o4-mini-deep-research, [+ 170 alpha/internal/legacy models]

Gemini (50 models — text/multimodal focused)

antigravity-preview-05-2026, gemini-2.0-flash, gemini-2.0-flash-001,
gemini-2.0-flash-lite, gemini-2.0-flash-lite-001,
gemini-2.5-computer-use-preview-10-2025, gemini-2.5-flash,
gemini-2.5-flash-image, gemini-2.5-flash-lite,
gemini-2.5-flash-native-audio-latest, gemini-2.5-flash-preview-tts,
gemini-2.5-pro, gemini-2.5-pro-preview-tts, gemini-3-flash-preview,
gemini-3-pro-image-preview, gemini-3-pro-preview,
gemini-3.1-flash-image-preview, gemini-3.1-flash-lite,
gemini-3.1-flash-lite-preview, gemini-3.1-flash-tts-preview,
gemini-3.1-pro-preview, gemini-3.1-pro-preview-customtools,
gemini-3.5-flash, gemini-flash-latest, gemini-flash-lite-latest,
gemini-pro-latest, deep-research-{max,}preview-04-2026,
deep-research-pro-preview-12-2025, gemma-4-26b-a4b-it, gemma-4-31b-it,
[+ imagen, veo, lyria, robotics, aqa, nano-banana-pro preview models]

Notes

• claude-opus-4-7 vs claude-opus-4.7: The Anthropic API returns claude-opus-4-7 (dashes), while model_multipliers.json stores claude-opus-4.7 (dot notation) with the correct 27.0 multiplier. Consider adding the dash-variant key as an alias entry.
• Legacy GPT models in Copilot: gpt-3.5-turbo, gpt-4, gpt-4-0613, gpt-4-o-preview are present in the Copilot endpoint but have no billing docs table entry — these appear to be free-tier legacy models; no multiplier entries are proposed.
• Gemini specialty models (video: veo-3.x, audio: lyria-3, image: imagen-4, robotics): These are non-text generative models without ET multipliers in the docs table; no aliases or multiplier entries proposed.
• gemini-flash alias cost warning: The broad gemini-flash alias now matches gemini-3.5-flash (14.0 multiplier) alongside cheap 0.1–0.2 flash variants. Workflow authors should use gemini-3.5-flash explicitly when targeting that model.

Generated by 📦 Daily Model Inventory Checker · sonnet46 1.6M · ◷

• expires on Jun 1, 2026, 12:19 AM UTC