feat: add SiliconCloud provider support #8552
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Add SiliconCloud provider implementation based on OpenAI-compatible API - Support multiple API endpoints (China, China-overseas, International) - Add model definitions with region-specific pricing - Integrate with UI settings and validation - Add i18n translations for SiliconCloud settings Closes #8550
- Extend BaseOpenAiCompatibleProvider for consistency - Fix type safety issues - Add proper imports
dosubot
bot
added
the
size:L
2 tasks
hannesrudolph
added
the
Issue/PR - Triage
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=60&v=4){kind=small}
Comment on lines
+56
to
+66
| const temperature = this.options.modelTemperature ?? SILICONCLOUD_DEFAULT_TEMPERATURE | ||
|
|
||
| // Build the request parameters | ||
| const params: OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming = { | ||
| model, | ||
| max_tokens, | ||
| temperature, | ||
| messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)], | ||
| stream: true, | ||
| stream_options: { include_usage: true }, | ||
| } |
Author
There was a problem hiding this comment.
[P1] Max tokens override from settings is ignored. Respect modelMaxTokens when present so the ThinkingBudget UI actually applies.
suggestion
Suggested change
| const temperature = this.options.modelTemperature ?? SILICONCLOUD_DEFAULT_TEMPERATURE | |
| // Build the request parameters | |
| const params: OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming = { | |
| model, | |
| max_tokens, | |
| temperature, | |
| messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)], | |
| stream: true, | |
| stream_options: { include_usage: true }, | |
| } | |
| const temperature = this.options.modelTemperature ?? SILICONCLOUD_DEFAULT_TEMPERATURE | |
| const resolved_max_tokens = this.options.modelMaxTokens ?? max_tokens | |
| // Build the request parameters | |
| const params: OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming = { | |
| model, | |
| max_tokens: resolved_max_tokens, | |
| temperature, | |
| messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)], | |
| stream: true, | |
| stream_options: { include_usage: true }, | |
| } |
Comment on lines
+71
to
+75
| if (supportsReasoningBudget) { | ||
| // SiliconCloud uses different parameter names than OpenAI | ||
| ;(params as any).enable_thinking = true | ||
| // Default thinking budget could be added here if needed | ||
| } |
Author
There was a problem hiding this comment.
[P1] SiliconCloud reasoning params: you set enable_thinking but never send thinking_budget and ignore enableReasoningEffort. This breaks budget control for reasoning-capable models.
suggestion
Suggested change
| if (supportsReasoningBudget) { | |
| // SiliconCloud uses different parameter names than OpenAI | |
| ;(params as any).enable_thinking = true | |
| // Default thinking budget could be added here if needed | |
| } | |
| if (supportsReasoningBudget && (this.options.enableReasoningEffort ?? true)) { | |
| // SiliconCloud uses different parameter names than OpenAI | |
| ;(params as any).enable_thinking = true | |
| const budget = this.options.modelMaxThinkingTokens ?? Math.floor(0.8 * ((params as any).max_tokens || 8192)) | |
| ;(params as any).thinking_budget = budget | |
| } |
Comment on lines
+226
to
+229
| // Export all models for the default list | ||
| export const siliconCloudModels = siliconCloudChinaModels | ||
|
|
||
| export type SiliconCloudModelId = keyof typeof siliconCloudModels |
Author
There was a problem hiding this comment.
[P2] Type-safety/UX: SiliconCloudModelId only includes China models, but you expose different model sets per apiLine. Unify the default list so static selectors and types cover all regions.
suggestion
Suggested change
| // Export all models for the default list | |
| export const siliconCloudModels = siliconCloudChinaModels | |
| export type SiliconCloudModelId = keyof typeof siliconCloudModels | |
| // Export all models for the default list (union across regions) | |
| export const siliconCloudModels = { | |
| ...siliconCloudChinaModels, | |
| ...siliconCloudChinaOverseasModels, | |
| ...siliconCloudInternationalModels, | |
| } | |
| export type SiliconCloudModelId = keyof typeof siliconCloudModels |
| export { Featherless } from "./Featherless" | ||
| export { VercelAiGateway } from "./VercelAiGateway" | ||
| export { DeepInfra } from "./DeepInfra" | ||
| export { SiliconCloud } from "./SiliconCloud" |
Author
There was a problem hiding this comment.
[P0] UI integration gap: SiliconCloud is exported here but not wired into the Settings UI. Without adding it to PROVIDERS and MODELS_BY_PROVIDER (webview-ui/src/components/settings/constants.ts) and rendering in ApiOptions (webview-ui/src/components/settings/ApiOptions.tsx), users cannot select or configure this provider. Please add:
- constants.ts: include { value: "siliconcloud", label: "SiliconCloud" } in PROVIDERS and map siliconcloud -> siliconCloudModels in MODELS_BY_PROVIDER
- ApiOptions.tsx: render when selectedProvider === "siliconcloud"
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Related GitHub Issue
Closes: #8550
Description
This PR implements a new SiliconCloud provider based on the OpenAI compatible provider, addressing the issues mentioned in #8550.
Key features include:
enable_thinkingandthinking_budget)Implementation Details
Test Procedure
Manual testing steps:
Notes
Pre-Submission Checklist
Important
Adds support for SiliconCloud provider with API handler, model definitions, UI components, and validation logic.
SiliconCloudHandlerinsrc/api/providers/siliconcloud.tsto handle SiliconCloud API interactions.enable_thinking,thinking_budget).packages/types/src/providers/siliconcloud.tswith region-specific pricing and context window information.webview-ui/src/components/ui/hooks/useSelectedModel.ts.SiliconCloudcomponent inwebview-ui/src/components/settings/providers/SiliconCloud.tsxfor UI settings.webview-ui/src/i18n/locales/en/settings.jsonfor SiliconCloud settings.validateApiConfigurationinwebview-ui/src/utils/validate.tsto include SiliconCloud-specific validation.ProfileValidatorinsrc/shared/ProfileValidator.ts.provider-settings.tsto include SiliconCloud in provider schemas and settings.This description was created by
for fc8f478. You can customize this summary. It will automatically update as commits are pushed.