[model-inventory] Model alias inventory - no changes needed - 2026-08-03

[github-actions]

Summary

Daily inventory check found the built-in model alias mapping (pkg/workflow/data/model_aliases.json) and cached pricing data (pkg/cli/data/models.json / actions/setup/js/models.json) are fully up to date. No changes proposed.

• Providers queried: OpenAI, Anthropic, Gemini, Copilot SDK, Copilot (reflect), Copilot billing docs
• Total live Copilot models: 36 (reflect) / 22 (SDK)
• Proposed alias changes: 0
• Pricing gaps found: 0

Provider Model Counts

Provider	Models Available	Status
openai	221	✅ ok
anthropic	10	✅ ok
gemini	50	✅ ok
copilot-sdk	22	✅ ok
copilot (reflect)	36	✅ ok
copilot billing docs	14 pricing rows	✅ ok

Alias Coverage Check

All 36 live Copilot model IDs from the reflect endpoint (claude-*, gpt-5.6-*, gpt-5.5, gpt-5.4*, gpt-5.3-codex, gpt-5-mini, mai-code-1-flash-picker, gemini-3.*, grok-4.5, deprecated gpt-3.5-turbo/gpt-4/gpt-4o*/embedding IDs, etc.) match at least one existing wildcard pattern in model_aliases.json (e.g. copilot/*sonnet*, copilot/gpt-5.6*, copilot/gemini-*pro*, copilot/*grok*). No uncovered live model IDs were found.

Newly seen dated/alpha OpenAI variants (e.g. gpt-5.4-2026-03-05, gpt-5.5-pro-2026-04-23, o3-2025-04-16) are development snapshots, not distinct families — they are already matched by the existing gpt-5*/gpt-5.4*/gpt-5-pro/reasoning wildcard patterns.

models.json Pricing Analysis

Cross-checked all 21 github-copilot entries in models.json against Copilot SDK billing.tokenPrices (per-100M-token units) and reflect billing/pricing.default fields, and against the docs pricing table (GPT-5 family). After normalizing units, every entry's input/output/cache_read cost matches the live SDK and docs data within rounding tolerance — no discrepancies found.

Missing from models.json

None — the one SDK-only entry (auto) is a routing alias, not a billable model, so it is intentionally excluded from the pricing table.

Historical entries not currently returned

claude-fable-5, claude-opus-4.5, claude-sonnet-4, gemini-2.5-pro, gemini-3-flash-preview, gpt-5.2, gpt-5.2-codex, gpt-5.4-nano, kimi-k2.7-code, raptor-mini — retained as historical records per policy (not currently in Copilot SDK/reflect live inventory, but no evidence they should be removed).

Inferred vs stored pricing discrepancies

None found.

Proposed Alias Updates

None. Existing aliases (sonnet, sonnet-6x, haiku, opus, gpt-5* family, gemini-* family, coding, summarization, reasoning, vision, task-oriented aliases, etc.) already provide complete coverage for all currently live models across all providers.

Notes

• gpt-4o-mini, gpt-4.1, gpt-4o, and gpt-5.4-nano were treated as intentionally deprecated per instructions and excluded from all gap/discrepancy analysis.
• Copilot billing docs table (fetched live via Playwright) cross-checks cleanly against SDK/reflect pricing for the GPT-5 family — no divergence.
• models.dev/api.json was reviewed as a secondary source; no additional gaps surfaced beyond the above.

Warning

Firewall blocked 7 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
• 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"
    - "safebrowsingohttpgateway.googleapis.com"
    - "www.google.com"

See Network Configuration for more information.

Generated by 📦 Daily Model Inventory Checker · auto · 76.7 AIC · ⌖ 7.22 AIC · ⊞ 12.6K · ◷

• expires on Aug 9, 2026, 4:22 PM UTC-08:00