fix: ensure OpenAI-compatible providers use custom max tokens setting

The issue was that OpenAI-compatible providers (Chutes, Groq) were directly using model.info.maxTokens instead of calling getModelMaxOutputTokens(). This meant that the user's custom modelMaxTokens setting was being ignored.

Fixed by:
- Updating BaseOpenAiCompatibleProvider to use getModelMaxOutputTokens()
- Updating ChutesHandler's getCompletionParams to use getModelMaxOutputTokens()

This ensures that when users set a custom max output tokens value in the settings, it will be properly applied to API requests for all OpenAI-compatible providers.

Author: MuriloFP

src/api/providers/base-openai-compatible-provider.ts

@@ -4,6 +4,7 @@ import OpenAI from "openai"
 import type { ModelInfo } from "@roo-code/types"
 
 import type { ApiHandlerOptions } from "../../shared/api"
+import { getModelMaxOutputTokens } from "../../shared/api"
 import { ApiStream } from "../transform/stream"
 import { convertToOpenAiMessages } from "../transform/openai-format"
 
@@ -67,15 +68,17 @@ export abstract class BaseOpenAiCompatibleProvider<ModelName extends string>
 		messages: Anthropic.Messages.MessageParam[],
 		metadata?: ApiHandlerCreateMessageMetadata,
 	): ApiStream {
-		const {
-			id: model,
-			info: { maxTokens: max_tokens },
-		} = this.getModel()
+		const model = this.getModel()
+		const max_tokens = getModelMaxOutputTokens({
+			modelId: model.id,
+			model: model.info,
+			settings: this.options as any,
+		})
 
 		const temperature = this.options.modelTemperature ?? this.defaultTemperature
 
 		const params: OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming = {
-			model,
+			model: model.id,
 			max_tokens,
 			temperature,
 			messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)],

src/api/providers/chutes.ts

@@ -3,6 +3,7 @@ import { Anthropic } from "@anthropic-ai/sdk"
 import OpenAI from "openai"
 
 import type { ApiHandlerOptions } from "../../shared/api"
+import { getModelMaxOutputTokens } from "../../shared/api"
 import { XmlMatcher } from "../../utils/xml-matcher"
 import { convertToR1Format } from "../transform/r1-format"
 import { convertToOpenAiMessages } from "../transform/openai-format"
@@ -27,15 +28,17 @@ export class ChutesHandler extends BaseOpenAiCompatibleProvider<ChutesModelId> {
 		systemPrompt: string,
 		messages: Anthropic.Messages.MessageParam[],
 	): OpenAI.Chat.Completions.ChatCompletionCreateParamsStreaming {
-		const {
-			id: model,
-			info: { maxTokens: max_tokens },
-		} = this.getModel()
+		const model = this.getModel()
+		const max_tokens = getModelMaxOutputTokens({
+			modelId: model.id,
+			model: model.info,
+			settings: this.options as any,
+		})
 
-		const temperature = this.options.modelTemperature ?? this.getModel().info.temperature
+		const temperature = this.options.modelTemperature ?? model.info.temperature
 
 		return {
-			model,
+			model: model.id,
 			max_tokens,
 			temperature,
 			messages: [{ role: "system", content: systemPrompt }, ...convertToOpenAiMessages(messages)],