[model-inventory] Model alias inventory update - 2026-07-27

[github-actions]

Summary

3 changes applied — claude-opus-5 added to models.json, gemini-3.6-flash alias added, agent alias updated to include gpt-5.5 and gpt-5.6.

• Providers queried: OpenAI (219 models), Anthropic (10 models), Gemini (50 models), Copilot (37 models via reflect)
• Total models found: ~316 across providers
• Proposed alias changes: 2 (1 new alias, 1 updated)
• Pricing gaps found: 1 (claude-opus-5 newly added)

Provider Model Counts

Provider	Models Available	Status
openai	219	✅ ok (API key configured)
anthropic	10	✅ ok (API key configured)
gemini	50	✅ ok (API key configured)
copilot	37	✅ ok (reflect endpoint)

Raw API Fields Discovered

Copilot (reflect): billing.multiplier is null for all models (no pricing multipliers exposed via reflect). Capabilities include max_context_window_tokens, max_output_tokens, max_prompt_tokens, supports.vision, supports.streaming, supports.tool_calls, supports.structured_outputs, supports.reasoning_effort, supports.adaptive_thinking.

Anthropic: display_name, created_at, type. No pricing fields in API response — pricing from models.dev.

Gemini: inputTokenLimit, outputTokenLimit, supportedGenerationMethods. Useful as proxy for pricing tier inference.

OpenAI: owned_by, created timestamp. Model ID naming conventions (e.g. -mini, -nano, -pro, -codex) used for tier inference.

models.dev: Full pricing (cost.input, cost.output, cost.cache_read, cost.cache_write), context limits, capability flags (reasoning, tool_call, structured_output, attachment), knowledge cutoff, release date.

Pricing Docs (GitHub Copilot billing page)

Extracted from https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing:

Model	Category	Input/1M	Output/1M
GPT-5 mini	Lightweight	$0.25	$2.00
GPT-5.3-Codex	Powerful	$1.75	$14.00
GPT-5.4	Versatile	$2.50	$15.00
GPT-5.4 mini	Lightweight	$0.75	$4.50
GPT-5.5	Powerful	$5.00	$30.00
GPT-5.6 Luna	Lightweight	$1.00	$6.00
GPT-5.6 Sol	Powerful	$5.00	$30.00
GPT-5.6 Terra	Versatile	$2.50	$15.00

Copilot billing.multiplier is null for all models — no ET multiplier data available from reflect endpoint.

models.json Pricing Analysis

Missing from models.json (now added)

Model ID	Provider	Basis
claude-opus-5	github-copilot / anthropic	models.dev catalog — $5/1M in, $25/1M out

Historical entries not currently returned by live Copilot

These remain in models.json as historical records:

Model ID	Note
claude-fable-5	Not in live Copilot reflect; kept in models.dev
claude-sonnet-4	Superseded by claude-sonnet-4.5+
gemini-3.1-pro-preview	Not currently in live Copilot
gemini-3.5-flash	Not currently in live Copilot
gpt-5.2	Older generation, not in live Copilot
gpt-5.2-codex	Older generation, not in live Copilot
kimi-k2.7-code	Not in live Copilot
raptor-mini	Not in live Copilot

Inferred vs stored pricing discrepancies

All existing entries validated against docs pricing table — no discrepancies found for models present in both.

Proposed Alias Updates (Applied)

1. New alias: gemini-3.6-flash

What: New specific version alias for the gemini-3.6-flash model now live in the Gemini API.
Why: Pattern parity with gemini-3.5-flash, gemini-3.1-flash etc. The generic gemini-flash wildcard covers it, but a specific alias allows explicit version pinning.
Syntax:

{
  "gemini-3.6-flash": [
    "copilot/gemini-3.6*flash*",
    "google/gemini-3.6*flash*",
    "gemini/gemini-3.6*flash*"
  ]
}

2. Updated alias: agent — add gpt-5.5 and gpt-5.6

What: Added gpt-5.5 and gpt-5.6 to the agent alias resolution order, inserted after gpt-5.4.
Why: gpt-5.5 (Powerful tier, $5/$30) and gpt-5.6-sol/-terra are now live Copilot models suitable for agent tasks. The previous agent alias only covered up to gpt-5.4 and gpt-5.3, leaving newer powerful generations unreachable via agent.
Syntax:

{
  "agent": [
    "sonnet-6x",
    "gpt-5.4",
    "gpt-5.5",
    "gpt-5.6",
    "gpt-5.3",
    "gemini-pro",
    "any"
  ]
}

Full Model Lists by Provider

Anthropic (10 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-opus-5,
claude-sonnet-4-5-20250929, claude-sonnet-4-6, claude-sonnet-5

Gemini (50 models, selection):

gemini-2.5-flash, gemini-2.5-flash-lite, gemini-2.5-pro, gemini-3-flash-preview,
gemini-3-pro-preview, gemini-3.1-flash-lite, gemini-3.1-pro-preview, gemini-3.5-flash,
gemini-3.5-flash-lite, gemini-3.6-flash (NEW), gemini-omni-flash-preview,
anti gravity-preview-05-2026, nano-banana-pro-preview, veo-3.1-generate-preview, lyria-3-pro-preview

OpenAI (219 models, notable):

gpt-5, gpt-5-codex, gpt-5-mini, gpt-5-nano, gpt-5-pro,
gpt-5.1, gpt-5.1-codex, 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, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra

Copilot (37 models via reflect):

claude-haiku-4.5, claude-opus-4.5, claude-opus-4.6, claude-opus-4.7, claude-opus-4.8,
claude-opus-5 (NEW), claude-sonnet-4.5, claude-sonnet-4.6, claude-sonnet-5,
gemini-2.5-pro, gemini-3-flash-preview, gpt-5-mini, gpt-5.3-codex, gpt-5.4,
gpt-5.4-mini, gpt-5.5, gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra,
mai-code-1-flash-picker, trajectory-compaction

Notes

• billing.multiplier is null for all Copilot models — no ET multiplier data available from the reflect endpoint for this run.
• gpt-4o-mini, gpt-4.1, gpt-4o, and gpt-5.4-nano are intentionally excluded as deprecated Copilot-facing IDs.
• Legacy embedding models (text-embedding-3-small, text-embedding-ada-002) and routing models (trajectory-compaction) appearing in the live Copilot reflect are not added to models.json — they serve infrastructure purposes.
• claude-opus-5 pricing validated against models.dev ($5/1M input, $25/1M output, $0.5/1M cache read) and is consistent with the Anthropic Opus family tier.

Warning

Firewall blocked 8 domains

The following domains were blocked by the firewall during workflow execution:

• accounts.google.com
• android.clients.google.com
• clients2.google.com
• collector.githubapp.com
• contentautofill.googleapis.com
• proxy.golang.org
• safebrowsingohttpgateway.googleapis.com
• www.google.com

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

network:
  allowed:
    - defaults
    - "accounts.google.com"
    - "android.clients.google.com"
    - "clients2.google.com"
    - "collector.githubapp.com"
    - "contentautofill.googleapis.com"
    - "proxy.golang.org"
    - "safebrowsingohttpgateway.googleapis.com"
    - "www.google.com"

See Network Configuration for more information.

Generated by 📦 Daily Model Inventory Checker · sonnet46 · 133.8 AIC · ⌖ 12.4 AIC · ⊞ 8.8K · ◷

• expires on Aug 2, 2026, 4:23 PM UTC-08:00