feat: Add provider-scoped router model fetches in frontend #8917

daniel-lxs · 2025-10-29T20:20:35Z

Problem

Part of the webview memory mitigation effort. The frontend was requesting all router models regardless of which provider was selected. This caused unnecessary memory allocation and cache reconciliation for models that would never be used.

Solution

Implement intelligent provider filtering in the frontend:

Identify which providers require router models (dynamic providers)
Disable useRouterModels entirely for static providers
For dynamic providers, request only the specific provider's models

Changes

useSelectedModel.ts

Added DYNAMIC_ROUTER_PROVIDERS set defining providers that need router models
Static providers (anthropic, openai-native, etc.): enabled: false
Dynamic providers: providers: [selectedProvider]

useRouterModels.ts

Added enabled option support
Added console.debug logging for observability
Enhanced provider filtering with request/response counters

Tests

Updated useSelectedModel.spec.ts for new static provider behavior

Impact

Before:

Fetched all router models (openrouter, roo, litellm, vercel, deepinfra, etc.) regardless of selected provider
Static providers like anthropic triggered unnecessary fetches

After:

Static providers: no router model fetches at all
Dynamic providers: only fetch the selected provider's models
Roo provider: only fetches Roo models

Memory savings:

Static provider users: ~8+ provider catalogs eliminated
Dynamic provider users: 7-8 provider catalogs eliminated (fetch only 1)

Files Changed

webview-ui/src/components/ui/hooks/useSelectedModel.ts
webview-ui/src/components/ui/hooks/useRouterModels.ts
webview-ui/src/components/ui/hooks/__tests__/useSelectedModel.spec.ts

Testing

✅ Updated tests for static provider behavior
✅ All webview UI tests passing
Console logs show reduced request frequency and narrower payloads

Part of

This is the third PR in a series to reduce webview memory usage:

✅ PR fix: Gate auth-driven Roo model refresh to active provider only #8915 - Gate auth-driven Roo model refresh
✅ PR feat: Add provider filtering support to router models backend #8916 - Backend provider filtering
This PR - Frontend provider-scoped fetches

Dependencies

Requires PR #8916 to be merged first for backend filtering support. This PR sends provider filters that the backend must honor.

Manual Testing

Select static provider (e.g., anthropic) → No router model requests in console
Select Roo provider → Console shows [useRouterModels] request providers=["roo"]
Switch providers → Verify only relevant provider models requested

Important

Optimizes router model fetching by introducing provider-specific logic, reducing unnecessary memory usage for static providers.

Behavior:
- useSelectedModel.ts: Introduces DYNAMIC_ROUTER_PROVIDERS set to identify providers needing router models.
- Disables useRouterModels for static providers and fetches only specific provider models for dynamic ones.
- Updates loading and error state logic to consider only necessary data.
Functions:
- useRouterModels.ts: Adds enabled option and provider filtering to getRouterModels().
- Adds request/response logging for observability.
Tests:
- Updates useSelectedModel.spec.ts to test new static provider behavior and ensure correct loading/error states.

^{This description was created by}^{for 1ff5392. You can customize this summary. It will automatically update as commits are pushed.}

Implements intelligent provider filtering in the frontend to request only the models needed for the currently selected provider. This dramatically reduces memory usage by avoiding unnecessary fetches and smaller cache footprints. - Add DYNAMIC_ROUTER_PROVIDERS set to identify providers needing models - Disable useRouterModels for static providers (anthropic, openai-native, etc.) - Pass provider-specific filters for dynamic providers - Add observability logging to track request/response patterns - Update tests for new behavior Works with backend filtering (PR #8916) to achieve end-to-end payload reduction.

roomote · 2025-10-29T20:21:01Z

Review Complete

No issues found. The implementation correctly:

Identifies static vs dynamic providers that need router model fetches
Gates useRouterModels entirely for static providers using the enabled option
Requests only provider-specific models for dynamic providers
Handles loading and error states properly by only considering relevant data sources
Updates tests to verify the new gating behavior
Guards against race conditions when multiple concurrent router model requests are in flight

Follow Along on Roo Code Cloud

…er filter to avoid race with active-only payload (pairs with #8916 backend filtering)

mrubens · 2025-10-31T02:02:58Z

webview-ui/src/components/ui/hooks/useSelectedModel.ts

 import { useLmStudioModels } from "./useLmStudioModels"
 import { useOllamaModels } from "./useOllamaModels"

+const DYNAMIC_ROUTER_PROVIDERS = new Set<ProviderName>([


Seems like we’re going to forget to update this - isn’t this already in the types somewhere?

mrubens · 2025-10-31T02:04:00Z

webview-ui/src/components/ui/hooks/useRouterModels.ts


 				if (message.routerModels) {
+					const keys = Object.keys(message.routerModels || {})
+					console.debug(


Can we remove this debug logging?

- Simplify backend to use single provider filter instead of array - Use existing isDynamicProvider from @roo-code/types instead of duplicating - Update frontend to request only the needed provider - Gate router model fetches entirely for static providers - Remove unnecessary debug logging - Update tests to reflect new behavior This combines and improves upon PRs #8916 and #8917 by: 1. Using a simpler single-provider API (provider: string instead of providers: string[]) 2. Leveraging existing type definitions from packages/types 3. Eliminating race conditions through simpler request/response matching 4. Reducing payload sizes and network usage Closes #8916 Closes #8917

daniel-lxs · 2025-10-31T16:35:23Z

Closing in favor of combined PR #8956 which addresses all feedback and uses a simpler single-provider API

daniel-lxs requested review from cte, jr and mrubens as code owners October 29, 2025 20:20

github-project-automation bot added this to Roo Code Roadmap Oct 29, 2025

github-project-automation bot moved this to Triage in Roo Code Roadmap Oct 29, 2025

github-project-automation bot added this to Roo Code Roadmap Oct 29, 2025

github-project-automation bot moved this to New in Roo Code Roadmap Oct 29, 2025

dosubot bot added the size:L This PR changes 100-499 lines, ignoring generated files. label Oct 29, 2025

daniel-lxs mentioned this pull request Oct 29, 2025

fix: Gate Roo router model fetches to reduce webview memory usage #8911

Closed

dosubot bot added the enhancement New feature or request label Oct 29, 2025

roomote bot approved these changes Oct 29, 2025

View reviewed changes

hannesrudolph added the Issue/PR - Triage New issue. Needs quick review to confirm validity and assign labels. label Oct 29, 2025

daniel-lxs moved this from Triage to PR [Needs Prelim Review] in Roo Code Roadmap Oct 29, 2025

hannesrudolph added PR - Needs Preliminary Review and removed Issue/PR - Triage New issue. Needs quick review to confirm validity and assign labels. labels Oct 29, 2025

ui(router-models): guard subscription to only resolve matching provid…

1ff5392

…er filter to avoid race with active-only payload (pairs with #8916 backend filtering)

roomote bot approved these changes Oct 30, 2025

View reviewed changes

daniel-lxs mentioned this pull request Oct 30, 2025

Router models: Only fetch models for the active provider #8912

Closed

daniel-lxs moved this from PR [Needs Prelim Review] to PR [Needs Review] in Roo Code Roadmap Oct 31, 2025

hannesrudolph added PR - Needs Review and removed PR - Needs Preliminary Review labels Oct 31, 2025

mrubens reviewed Oct 31, 2025

View reviewed changes

daniel-lxs mentioned this pull request Oct 31, 2025

feat: optimize router model fetching with single-provider filtering #8956

Merged

daniel-lxs closed this Oct 31, 2025

github-project-automation bot moved this from New to Done in Roo Code Roadmap Oct 31, 2025

github-project-automation bot moved this from PR [Needs Review] to Done in Roo Code Roadmap Oct 31, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: Add provider-scoped router model fetches in frontend #8917

feat: Add provider-scoped router model fetches in frontend #8917

Uh oh!

daniel-lxs commented Oct 29, 2025 •

edited by ellipsis-dev bot

Loading

Uh oh!

roomote bot commented Oct 29, 2025 •

edited

Loading

Uh oh!

mrubens Oct 31, 2025

Uh oh!

mrubens Oct 31, 2025

Uh oh!

daniel-lxs commented Oct 31, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

feat: Add provider-scoped router model fetches in frontend #8917

feat: Add provider-scoped router model fetches in frontend #8917

Uh oh!

Conversation

daniel-lxs commented Oct 29, 2025 • edited by ellipsis-dev bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Problem

Solution

Changes

useSelectedModel.ts

useRouterModels.ts

Tests

Impact

Files Changed

Testing

Part of

Dependencies

Manual Testing

Uh oh!

roomote bot commented Oct 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review Complete

Uh oh!

mrubens Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

mrubens Oct 31, 2025

Choose a reason for hiding this comment

Uh oh!

daniel-lxs commented Oct 31, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

daniel-lxs commented Oct 29, 2025 •

edited by ellipsis-dev bot

Loading

roomote bot commented Oct 29, 2025 •

edited

Loading