[model-inventory] Model alias inventory update - 2026-07-04

[github-actions]

Summary

Daily model alias inventory check for 2026-07-04.

• Providers queried: OpenAI, Anthropic, Gemini, Copilot
• Total models found: 308 (Copilot: 32, Anthropic: 9, OpenAI: 217, Gemini: 50)
• Proposed alias changes: 2 (new lyria and audio aliases)
• Pricing gaps found: 1 (openai/gpt-5.6 missing from models.json)

Provider Model Counts

Provider	Models Available	Status
copilot	32	✅ ok
anthropic	9	✅ ok
openai	217	✅ ok
gemini	50	✅ ok

Raw API Fields Discovered

Anthropic — max_input_tokens, max_tokens, capabilities.batch, capabilities.effort (low/medium/high/xhigh/max), capabilities.thinking, capabilities.image_input, capabilities.pdf_input, capabilities.structured_outputs, capabilities.code_execution, capabilities.context_management (compact, clear_tool_uses, clear_thinking). Noteworthy: effort.xhigh is false for sonnet-4.6 but true for sonnet-5 and opus-4.x.

OpenAI — id, object, created, owned_by. Minimal metadata; model family and tier inferred from ID naming conventions. API mapping table in reflect (model_api_mapping) indicates Responses-API-only for GPT-5.x and Chat-Completions for GPT-4.x and earlier.

Gemini — inputTokenLimit, outputTokenLimit, supportedGenerationMethods, temperature, topP, topK, maxTemperature, thinking flag, version, displayName, description. TTS models have small context (8192 tokens); standard models 1M tokens. Gemini 2.5 Flash-Image ("Nano Banana") has 32K context.

Copilot — /reflect returns billing.multiplier indirectly via models-dev pricing data (per-token cost fields). wire_api (responses / completions) and provider_type are authoritative routing fields for SDK selection. All current models have correct wire_api and provider_type entries.

models.json Pricing Analysis

Missing from models.json

Model ID	Provider	Inferred Pricing	Basis
gpt-5.6	openai	~$5/M input, $30/M output (inferred from gpt-5.5 tier, same OpenAI pricing tier)	OpenAI live API; no models-dev entry yet; gpt-5.6 alias already exists

Historical entries not currently returned

The following model IDs appear in pkg/cli/data/models.json but are absent from all live provider inventories queried. Treat as historical records — no action required:

Model ID	Section	Note
claude-fable-5	github-copilot	Not yet in reflect (preview/early access)
claude-sonnet-4	github-copilot	Previous generation
gemini-3.1-pro-preview	github-copilot	Not yet in reflect
gemini-3.5-flash	github-copilot	Not yet in reflect
gpt-5.2	github-copilot	Previous generation
gpt-5.2-codex	github-copilot	Previous generation
gpt-5.4-nano	github-copilot	Deprecated (intentional, see task instructions)
raptor-mini	github-copilot	Not yet in reflect
kimi-k2.7-code	github-copilot	Not yet in reflect

Inferred vs stored pricing discrepancies

No discrepancies found. All 25 entries in the github-copilot section of models.json match models-dev pricing exactly (within floating-point precision). All Copilot reflect billing data aligns with stored values.

Proposed Alias Updates

1. Add lyria alias

What: New alias for Google Lyria audio/music generation models.

Why: lyria-3-pro-preview and lyria-3-clip-preview appeared in the Gemini live API. These are dedicated audio/music generation models — a new family that has no specific alias. The veo alias provides a precedent for media-generation aliases.

Syntax:

{
  "lyria": [
    "google/lyria*",
    "gemini/lyria*"
  ]
}

2. Add audio alias

What: New semantic alias for audio generation/TTS models across providers.

Why: OpenAI has gpt-audio, gpt-audio-mini, gpt-audio-1.5; Gemini has lyria-3-* and TTS preview models. No unified audio alias exists. This mirrors image-generation and veo as media-task aliases.

Syntax:

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

Notes

• kimi and kiwi aliases have no live matches in copilot or OpenAI inventories — they may be stale, but both patterns are forward-looking (model IDs could appear at any time). No removal proposed; leave as-is.
• gemini-omni, gemma, fable, computer-use, robotics, veo — all confirmed matching at least one live model in the full Gemini/OpenAI live inventories; not stale.
• claude-sonnet-5 (new generation) is correctly covered by the existing sonnet wildcard alias copilot/*sonnet*.
• gpt-5.6 (new OpenAI generation with luna/sol/terra variants) is covered by both gpt-5 (broad) and gpt-5.6 (specific) aliases. Pricing entry in models.json openai section is missing but should be added when models-dev publishes it.
• All Copilot-routed models have correct wire_api and provider_type in models.json.

Full Model Lists by Provider

Copilot (32 models)

claude-haiku-4.5, claude-opus-4.5, claude-opus-4.6, claude-opus-4.7, claude-opus-4.8, claude-sonnet-4.5, claude-sonnet-4.6, claude-sonnet-5, gemini-2.5-pro, gemini-3-flash-preview, 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-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.5, mai-code-1-flash-picker, text-embedding-3-small, text-embedding-3-small-inference, text-embedding-ada-002

Anthropic API (9 models)

claude-haiku-4-5-20251001, claude-opus-4-1-20250805, claude-opus-4-5-20251101, claude-opus-4-6, claude-opus-4-7, claude-opus-4-8, claude-sonnet-4-5-20250929, claude-sonnet-4-6, claude-sonnet-5

OpenAI API (selected relevant models)

gpt-5, gpt-5-mini, gpt-5-nano, gpt-5-pro, gpt-5-codex, gpt-5.1, gpt-5.1-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2, gpt-5.2-codex, gpt-5.2-pro, gpt-5.3-codex, gpt-5.4, gpt-5.4-mini, gpt-5.4-pro, gpt-5.5, gpt-5.5-pro, gpt-5.6, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra, o1, o1-pro, o3, o3-pro, o4-mini, o3-deep-research, o4-mini-deep-research, gpt-image-1, gpt-image-2, computer-use-preview

Gemini API (selected relevant models)

gemini-2.0-flash, gemini-2.0-flash-lite, gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.5-pro, gemini-3-flash-preview, gemini-3-pro-preview, gemini-3-pro-image, gemini-3.1-flash-lite, gemini-3.1-pro-preview, gemini-3.5-flash, gemini-omni-flash-preview, antigravity-preview-05-2026, nano-banana-pro-preview, gemma-4-26b-a4b-it, gemma-4-31b-it, lyria-3-pro-preview, lyria-3-clip-preview, veo-3.1-generate-preview, veo-3.1-fast-generate-preview

Warning

Firewall blocked 1 domain

The following domain was blocked by the firewall during workflow execution:

• models.dev

To allow these domains, add them to the network.allowed list in your workflow frontmatter:

network:
  allowed:
    - defaults
    - "models.dev"

See Network Configuration for more information.

Generated by 📦 Daily Model Inventory Checker · 159.6 AIC · ⌖ 14.4 AIC · ⊞ 7.8K · ◷

• expires on Jul 10, 2026, 4:30 PM UTC-08:00

[github-actions]

This issue is being closed as outdated. A newer issue has been created: #43444

View newer issue

---

This action was performed automatically by the Daily Model Inventory Checker workflow.