Requesty: Add model info (RooCodeInc#2190)

Adding model information to Requesty provider.

- Add a new model picker component for Requesty.
- Enable controlling thinking budget via a slider

IMPORTANT:
Model information is fetched ONLY(!) when the user chooses "requesty"
as their provider to avoid any boot latency.

Author: dtrugman

.changeset/rude-apes-behave.md

@@ -0,0 +1,5 @@
+---
+"claude-dev": patch
+---
+
+Add model info to the Requesty provider

src/api/providers/requesty.ts

@@ -1,9 +1,10 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import OpenAI from "openai"
-import { ApiHandlerOptions, ModelInfo, openAiModelInfoSaneDefaults } from "../../shared/api"
+import { ApiHandlerOptions, ModelInfo, requestyDefaultModelId, requestyDefaultModelInfo } from "../../shared/api"
 import { ApiHandler } from "../index"
 import { withRetry } from "../retry"
 import { convertToOpenAiMessages } from "../transform/openai-format"
+import { calculateApiCostOpenAI } from "../../utils/cost"
 import { ApiStream } from "../transform/stream"
 
 export class RequestyHandler implements ApiHandler {
@@ -24,21 +25,34 @@ export class RequestyHandler implements ApiHandler {
 
 	@withRetry()
 	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
-		const modelId = this.options.requestyModelId ?? ""
+		const model = this.getModel()
 
 		const openAiMessages: OpenAI.Chat.ChatCompletionMessageParam[] = [
 			{ role: "system", content: systemPrompt },
 			...convertToOpenAiMessages(messages),
 		]
 
+		const reasoningEffort = this.options.o3MiniReasoningEffort || "medium"
+		const reasoning = { reasoning_effort: reasoningEffort }
+		const reasoningArgs = model.id === "openai/o3-mini" ? reasoning : {}
+
+		const thinkingBudget = this.options.thinkingBudgetTokens || 0
+		const thinking =
+			thinkingBudget > 0
+				? { thinking: { type: "enabled", budget_tokens: thinkingBudget } }
+				: { thinking: { type: "disabled" } }
+		const thinkingArgs = model.id.includes("claude-3-7-sonnet") ? thinking : {}
+
 		// @ts-ignore-next-line
 		const stream = await this.client.chat.completions.create({
-			model: modelId,
+			model: model.id,
+			max_tokens: model.info.maxTokens || undefined,
 			messages: openAiMessages,
 			temperature: 0,
 			stream: true,
 			stream_options: { include_usage: true },
-			...(modelId === "openai/o3-mini" ? { reasoning_effort: this.options.o3MiniReasoningEffort || "medium" } : {}),
+			...reasoningArgs,
+			...thinkingArgs,
 		})
 
 		for await (const chunk of stream) {
@@ -73,7 +87,7 @@ export class RequestyHandler implements ApiHandler {
 				const outputTokens = usage.completion_tokens || 0
 				const cacheWriteTokens = usage.prompt_tokens_details?.caching_tokens || undefined
 				const cacheReadTokens = usage.prompt_tokens_details?.cached_tokens || undefined
-				const totalCost = 0 // TODO: Replace with calculateApiCostOpenAI(model.info, inputTokens, outputTokens, cacheWriteTokens, cacheReadTokens)
+				const totalCost = calculateApiCostOpenAI(model.info, inputTokens, outputTokens, cacheWriteTokens, cacheReadTokens)
 
 				yield {
 					type: "usage",
@@ -88,9 +102,11 @@ export class RequestyHandler implements ApiHandler {
 	}
 
 	getModel(): { id: string; info: ModelInfo } {
-		return {
-			id: this.options.requestyModelId ?? "",
-			info: openAiModelInfoSaneDefaults,
+		const modelId = this.options.requestyModelId
+		const modelInfo = this.options.requestyModelInfo
+		if (modelId && modelInfo) {
+			return { id: modelId, info: modelInfo }
 		}
+		return { id: requestyDefaultModelId, info: requestyDefaultModelInfo }
 	}
 }

src/core/controller/index.ts

@@ -487,6 +487,9 @@ export class Controller {
 			case "refreshOpenRouterModels":
 				await this.refreshOpenRouterModels()
 				break
+			case "refreshRequestyModels":
+				await this.refreshRequestyModels()
+				break
 			case "refreshOpenAiModels":
 				const { apiConfiguration } = await getAllExtensionState(this.context)
 				const openAiModels = await this.getOpenAiModels(apiConfiguration.openAiBaseUrl, apiConfiguration.openAiApiKey)
@@ -983,6 +986,7 @@ export class Controller {
 					break
 				case "requesty":
 					await updateGlobalState(this.context, "previousModeModelId", apiConfiguration.requestyModelId)
+					await updateGlobalState(this.context, "previousModeModelInfo", apiConfiguration.requestyModelInfo)
 					break
 			}
 
@@ -1024,6 +1028,7 @@ export class Controller {
 						break
 					case "requesty":
 						await updateGlobalState(this.context, "requestyModelId", newModelId)
+						await updateGlobalState(this.context, "requestyModelInfo", newModelInfo)
 						break
 				}
 
@@ -1621,6 +1626,52 @@ Here is the project's README to help you get started:\n\n${mcpDetails.readmeCont
 		return models
 	}
 
+	async refreshRequestyModels() {
+		const parsePrice = (price: any) => {
+			if (price) {
+				return parseFloat(price) * 1_000_000
+			}
+			return undefined
+		}
+
+		let models: Record<string, ModelInfo> = {}
+		try {
+			const apiKey = await getSecret(this.context, "requestyApiKey")
+			const headers = {
+				Authorization: `Bearer ${apiKey}`,
+			}
+			const response = await axios.get("https://router.requesty.ai/v1/models", { headers })
+			if (response.data?.data) {
+				for (const model of response.data.data) {
+					const modelInfo: ModelInfo = {
+						maxTokens: model.max_output_tokens || undefined,
+						contextWindow: model.context_window,
+						supportsImages: model.supports_vision || undefined,
+						supportsComputerUse: model.supports_computer_use || undefined,
+						supportsPromptCache: model.supports_caching || undefined,
+						inputPrice: parsePrice(model.input_price),
+						outputPrice: parsePrice(model.output_price),
+						cacheWritesPrice: parsePrice(model.caching_price),
+						cacheReadsPrice: parsePrice(model.cached_price),
+						description: model.description,
+					}
+					models[model.id] = modelInfo
+				}
+				console.log("Requesty models fetched", models)
+			} else {
+				console.error("Invalid response from Requesty API")
+			}
+		} catch (error) {
+			console.error("Error fetching Requesty models:", error)
+		}
+
+		await this.postMessageToWebview({
+			type: "requestyModels",
+			requestyModels: models,
+		})
+		return models
+	}
+
 	// Context menus and code actions
 
 	getFileMentionFromPath(filePath: string) {

src/core/storage/state-keys.ts

@@ -61,6 +61,7 @@ export type GlobalStateKey =
 	| "liteLlmUsePromptCache"
 	| "qwenApiLine"
 	| "requestyModelId"
+	| "requestyModelInfo"
 	| "togetherModelId"
 	| "mcpMarketplaceCatalog"
 	| "telemetrySetting"

src/core/storage/state.ts

@@ -83,6 +83,7 @@ export async function getAllExtensionState(context: vscode.ExtensionContext) {
 		deepSeekApiKey,
 		requestyApiKey,
 		requestyModelId,
+		requestyModelInfo,
 		togetherApiKey,
 		togetherModelId,
 		qwenApiKey,
@@ -150,6 +151,7 @@ export async function getAllExtensionState(context: vscode.ExtensionContext) {
 		getSecret(context, "deepSeekApiKey") as Promise<string | undefined>,
 		getSecret(context, "requestyApiKey") as Promise<string | undefined>,
 		getGlobalState(context, "requestyModelId") as Promise<string | undefined>,
+		getGlobalState(context, "requestyModelInfo") as Promise<ModelInfo | undefined>,
 		getSecret(context, "togetherApiKey") as Promise<string | undefined>,
 		getGlobalState(context, "togetherModelId") as Promise<string | undefined>,
 		getSecret(context, "qwenApiKey") as Promise<string | undefined>,
@@ -256,6 +258,7 @@ export async function getAllExtensionState(context: vscode.ExtensionContext) {
 			deepSeekApiKey,
 			requestyApiKey,
 			requestyModelId,
+			requestyModelInfo,
 			togetherApiKey,
 			togetherModelId,
 			qwenApiKey,
@@ -329,6 +332,7 @@ export async function updateApiConfiguration(context: vscode.ExtensionContext, a
 		deepSeekApiKey,
 		requestyApiKey,
 		requestyModelId,
+		requestyModelInfo,
 		togetherApiKey,
 		togetherModelId,
 		qwenApiKey,
@@ -397,6 +401,7 @@ export async function updateApiConfiguration(context: vscode.ExtensionContext, a
 	await updateGlobalState(context, "liteLlmUsePromptCache", liteLlmUsePromptCache)
 	await updateGlobalState(context, "qwenApiLine", qwenApiLine)
 	await updateGlobalState(context, "requestyModelId", requestyModelId)
+	await updateGlobalState(context, "requestyModelInfo", requestyModelInfo)
 	await updateGlobalState(context, "togetherModelId", togetherModelId)
 	await storeSecret(context, "asksageApiKey", asksageApiKey)
 	await updateGlobalState(context, "asksageApiUrl", asksageApiUrl)

src/shared/ExtensionMessage.ts

@@ -24,6 +24,7 @@ export interface ExtensionMessage {
 		| "partialMessage"
 		| "openRouterModels"
 		| "openAiModels"
+		| "requestyModels"
 		| "mcpServers"
 		| "relinquishControl"
 		| "vsCodeLmModels"
@@ -69,6 +70,7 @@ export interface ExtensionMessage {
 	partialMessage?: ClineMessage
 	openRouterModels?: Record<string, ModelInfo>
 	openAiModels?: string[]
+	requestyModels?: Record<string, ModelInfo>
 	mcpServers?: McpServer[]
 	customToken?: string
 	mcpMarketplaceCatalog?: McpMarketplaceCatalog

src/shared/WebviewMessage.ts

@@ -30,6 +30,7 @@ export interface WebviewMessage {
 		| "openMention"
 		| "cancelTask"
 		| "refreshOpenRouterModels"
+		| "refreshRequestyModels"
 		| "refreshOpenAiModels"
 		| "openMcpSettings"
 		| "restartMcpServer"

src/shared/api.ts

@@ -60,6 +60,7 @@ export interface ApiHandlerOptions {
 	deepSeekApiKey?: string
 	requestyApiKey?: string
 	requestyModelId?: string
+	requestyModelInfo?: ModelInfo
 	togetherApiKey?: string
 	togetherModelId?: string
 	qwenApiKey?: string
@@ -1554,3 +1555,19 @@ export const sambanovaModels = {
 		outputPrice: 1.5,
 	},
 } as const satisfies Record<string, ModelInfo>
+
+// Requesty
+// https://requesty.ai/models
+export const requestyDefaultModelId = "anthropic/claude-3-7-sonnet-latest"
+export const requestyDefaultModelInfo: ModelInfo = {
+	maxTokens: 8192,
+	contextWindow: 200_000,
+	supportsImages: true,
+	supportsComputerUse: false,
+	supportsPromptCache: true,
+	inputPrice: 3.0,
+	outputPrice: 15.0,
+	cacheWritesPrice: 3.75,
+	cacheReadsPrice: 0.3,
+	description: "Anthropic's most intelligent model. Highest level of intelligence and capability.",
+}

webview-ui/src/components/settings/ApiOptions.tsx

@@ -32,6 +32,8 @@ import {
 	openAiNativeModels,
 	openRouterDefaultModelId,
 	openRouterDefaultModelInfo,
+	requestyDefaultModelId,
+	requestyDefaultModelInfo,
 	mainlandQwenModels,
 	internationalQwenModels,
 	mainlandQwenDefaultModelId,
@@ -56,6 +58,7 @@ import { getAsVar, VSC_DESCRIPTION_FOREGROUND } from "@/utils/vscStyles"
 import VSCodeButtonLink from "@/components/common/VSCodeButtonLink"
 import OpenRouterModelPicker, { ModelDescriptionMarkdown, OPENROUTER_MODEL_PICKER_Z_INDEX } from "./OpenRouterModelPicker"
 import { ClineAccountInfoCard } from "./ClineAccountInfoCard"
+import RequestyModelPicker from "./RequestyModelPicker"
 
 interface ApiOptionsProps {
 	showModelOptions: boolean
@@ -1050,24 +1053,7 @@ const ApiOptions = ({ showModelOptions, apiErrorMessage, modelIdErrorMessage, is
 						placeholder="Enter API Key...">
 						<span style={{ fontWeight: 500 }}>API Key</span>
 					</VSCodeTextField>
-					<VSCodeTextField
-						value={apiConfiguration?.requestyModelId || ""}
-						style={{ width: "100%" }}
-						onInput={handleInputChange("requestyModelId")}
-						placeholder={"Enter Model ID..."}>
-						<span style={{ fontWeight: 500 }}>Model ID</span>
-					</VSCodeTextField>
-					<p
-						style={{
-							fontSize: "12px",
-							marginTop: 3,
-							color: "var(--vscode-descriptionForeground)",
-						}}>
-						<span style={{ color: "var(--vscode-errorForeground)" }}>
-							(<span style={{ fontWeight: 500 }}>Note:</span> Cline uses complex prompts and works best with Claude
-							models. Less capable models may not work as expected.)
-						</span>
-					</p>
+					{!apiConfiguration?.requestyApiKey && <a href="https://app.requesty.ai/manage-api">Get API Key</a>}
 				</div>
 			)}
 
@@ -1561,6 +1547,7 @@ const ApiOptions = ({ showModelOptions, apiErrorMessage, modelIdErrorMessage, is
 			{(selectedProvider === "openrouter" || selectedProvider === "cline") && showModelOptions && (
 				<OpenRouterModelPicker isPopup={isPopup} />
 			)}
+			{selectedProvider === "requesty" && showModelOptions && <RequestyModelPicker isPopup={isPopup} />}
 
 			{modelIdErrorMessage && (
 				<p
@@ -1814,6 +1801,12 @@ export function normalizeApiConfiguration(apiConfiguration?: ApiConfiguration):
 				selectedModelId: apiConfiguration?.openRouterModelId || openRouterDefaultModelId,
 				selectedModelInfo: apiConfiguration?.openRouterModelInfo || openRouterDefaultModelInfo,
 			}
+		case "requesty":
+			return {
+				selectedProvider: provider,
+				selectedModelId: apiConfiguration?.requestyModelId || requestyDefaultModelId,
+				selectedModelInfo: apiConfiguration?.requestyModelInfo || requestyDefaultModelInfo,
+			}
 		case "cline":
 			return {
 				selectedProvider: provider,