fix: correct deprecated model settings behavior

[erichare]

Summary

• disable model-provider switches for deprecated and unsupported models
• reject API attempts to persist those models as explicitly enabled
• limit age-based auto-deprecation to language models so embeddings do not become deprecated solely because they cross the 900-day threshold

Root cause

The settings UI allowed deprecated models to enter its optimistic toggle queue, and the update endpoint accepted and persisted that state. A later GET /enabled_models always returned deprecated models as disabled, so the next refetch appeared to undo the user's toggle.

Separately, the models.dev fallback marks catalog entries older than 900 days as deprecated. That heuristic also applied to embeddings. The OpenAI text-embedding-3-* models reached that threshold in July 2026 even though age alone is not a reliable deprecation signal for embedding models.

Explicit catalog deprecations remain unchanged, including provider-specific embedding models that are known to be unavailable.

Validation

• uv run pytest tests/unit/base/models/test_models_dev_catalog.py -q (23 passed)
• uv run pytest src/backend/tests/unit/api/v1/test_models_enabled_providers.py -k cannot_enable_deprecated_model -q (1 passed)
• npm test -- src/modals/modelProviderModal/__tests__/ModelSelection.test.tsx --runInBand (26 passed)
• focused Ruff and Biome checks
• pre-commit hooks
• live models.dev smoke test confirms all three OpenAI embedding models remain non-deprecated

Summary by CodeRabbit

• New Features

  • Deprecated and unsupported models can no longer be enabled.
  • Model selection controls are disabled for unavailable models.
  • Clear error messages are shown when attempting to enable deprecated models.

• Bug Fixes

  • Age-based deprecation now applies only to language models; embeddings remain available unless explicitly deprecated.

• Tests

  • Added coverage for unavailable-model restrictions and credential cleanup behavior.

[erichare]

Replaces #14046 with the same commit on an upstream-owned branch.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: d764af4d-d472-4498-9335-f099a84bced5

📥 Commits

Reviewing files that changed from the base of the PR and between 61d6490 and 860e8db.

📒 Files selected for processing (7)

• src/backend/base/langflow/api/v1/models.py
• src/backend/tests/unit/api/v1/test_models_enabled_providers.py
• src/backend/tests/unit/api/v1/test_variable.py
• src/frontend/src/modals/modelProviderModal/__tests__/ModelSelection.test.tsx
• src/frontend/src/modals/modelProviderModal/components/ModelSelection.tsx
• src/lfx/src/lfx/base/models/models_dev_catalog.py
• src/lfx/tests/unit/base/models/test_models_dev_catalog.py

---

Walkthrough

Model enablement now reuses the model catalog, rejects deprecated or unsupported models in the backend, disables their frontend toggles, and clarifies language-model-only age-based deprecation behavior.

Changes

Model enablement safeguards

Layer / File(s)	Summary
Catalog deprecation semantics src/lfx/src/lfx/base/models/models_dev_catalog.py, src/lfx/tests/unit/base/models/test_models_dev_catalog.py	Documentation and tests clarify that age-based auto-deprecation applies only to language models while static embedding deprecation flags are preserved.
Backend unavailable-model validation src/backend/base/langflow/api/v1/models.py, src/backend/tests/unit/api/v1/test_models_enabled_providers.py, src/backend/tests/unit/api/v1/test_variable.py	The enabled-model endpoint reuses the fetched catalog, rejects deprecated or unsupported models, removes per-update credential validation, and updates related cleanup tests.
Frontend unavailable-model controls src/frontend/src/modals/modelProviderModal/components/ModelSelection.tsx, src/frontend/src/modals/modelProviderModal/__tests__/ModelSelection.test.tsx	Deprecated and unsupported model toggles are disabled and covered by a UI test.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Sequence Diagram(s)

sequenceDiagram
  participant Client
  participant ModelSelection
  participant update_enabled_models
  participant ModelCatalog
  Client->>ModelSelection: View model metadata
  ModelSelection->>ModelSelection: Disable deprecated or unsupported toggles
  Client->>update_enabled_models: Submit enabled model IDs
  update_enabled_models->>ModelCatalog: Fetch model catalog
  ModelCatalog-->>update_enabled_models: Return availability metadata
  update_enabled_models-->>Client: Reject unavailable model or update settings

Loading

Possibly related PRs

• langflow-ai/langflow#13191: Changes WatsonX deprecated metadata consumed by the enablement validation.
• langflow-ai/langflow#14023: Overlaps with the model catalog and enabled-model endpoint changes.

Suggested labels: lgtm

Suggested reviewers: himavarshavs, edwinjosechittilappilly, lucaseduoli, ramgopalsrikar, cristhianzl

🚥 Pre-merge checks | ✅ 9

✅ Passed checks (9 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title is concise and clearly matches the main change: fixing deprecated model settings behavior.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Test Coverage For New Implementations	✅ Passed	PASS: Added real regression tests for backend API, frontend toggle disabling, and LFX auto-deprecation; existing cleanup test updated, and filenames match project conventions.
Test Quality And Coverage	✅ Passed	Tests cover the new backend 400 path, supported-model 200 flow, frontend disabled-toggle behavior, and LFX age-based deprecation, using the repo’s async pytest/Jest patterns.
Test File Naming And Structure	✅ Passed	Changed tests match repo conventions: backend pytest unit files, a frontend .test.tsx Jest/RTL test with grouped describe/beforeEach blocks, and descriptive positive/negative cases.
Excessive Mock Usage Warning	✅ Passed	Mocking is scoped to external deps (API key/network/hook isolation); the new deprecated-model tests are integration-style and don’t add extra mocks.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/deprecated-model-settings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

✅ Test Coverage Advisor

No source changes detected without accompanying tests. Thanks for keeping coverage up! 🎉

Advisory check only — never blocks merge.

[erichare]

Release-1.11.0 companion: #14048

[codecov]

Codecov Report

❌ Patch coverage is 28.57143% with 20 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (release-1.10.3@61d6490). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
src/backend/base/langflow/api/v1/models.py	9.09%	20 Missing ⚠️

❌ Your patch status has failed because the patch coverage (28.57%) is below the target coverage (40.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@                Coverage Diff                @@
##             release-1.10.3   #14047   +/-   ##
=================================================
  Coverage                  ?   58.58%           
=================================================
  Files                     ?     2307           
  Lines                     ?   220276           
  Branches                  ?    32985           
=================================================
  Hits                      ?   129053           
  Misses                    ?    89737           
  Partials                  ?     1486           

Flag	Coverage Δ
backend	65.88% <9.09%> (?)
frontend	57.69% <100.00%> (?)
lfx	54.65% <100.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...s/modelProviderModal/components/ModelSelection.tsx	99.26% <100.00%> (ø)
src/lfx/src/lfx/base/models/models_dev_catalog.py	86.59% <100.00%> (ø)
src/backend/base/langflow/api/v1/models.py	39.79% <9.09%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Frontend Unit Test Coverage Report

Coverage Summary

Lines	Statements	Branches	Functions
	43.54% (58076/133366)	69.19% (7885/11395)	41.56% (1298/3123)

Unit Test Results

Tests	Skipped	Failures	Errors	Time
4979	0 💤	0 ❌	0 🔥	12m 48s ⏱️

[lucaseduoli]

LGTM!