feat(xai): add dynamic model discovery with correct pricing

- Implemented dynamic model discovery for xAI provider using /v1/language-models endpoint
- Models are now fetched at runtime with real-time pricing and capabilities
- Fixed pricing conversion: XAI API returns fractional cents (basis points), divide by 10,000 not 100
- This fixes pricing display showing $20.00 instead of $0.20 per 1M tokens
- Removed static xAI models from MODELS_BY_PROVIDER to rely on dynamic discovery
- Enhanced error logging with detailed status and URL information
- Support dynamic model context window overrides from API
- Fixed parseApiPrice to handle zero values correctly (for free models)
- Provided complete ModelInfo fallback in useSelectedModel for UI type safety
- Added comprehensive test coverage for cost utilities and XAI fetcher
- Updated all tests to reflect correct pricing scale and dynamic model behavior

Author: hannesrudolph

.changeset/v3.29.4.md

@@ -1,18 +1,5 @@
 # Roo Code Changelog
 
-## [3.29.3] - 2025-10-28
-
-- Update Gemini models with latest 09-2025 versions including Gemini 2.5 Pro and Flash (#8485 by @cleacos, PR by @roomote)
-- Add reasoning support for Z.ai GLM binary thinking mode (#8465 by @BeWater799, PR by @daniel-lxs)
-- Enable reasoning in Roo provider (thanks @mrubens!)
-- Add settings to configure time and cost display in system prompt (#8450 by @jaxnb, PR by @roomote)
-- Fix: Use max_output_tokens when available in LiteLLM fetcher (#8454 by @fabb, PR by @roomote)
-- Fix: Process queued messages after context condensing completes (#8477 by @JosXa, PR by @roomote)
-- Fix: Use monotonic clock for rate limiting to prevent timing issues (#7770 by @intermarkec, PR by @chrarnoldus)
-- Fix: Resolve checkpoint menu popover overflow (thanks @daniel-lxs!)
-- Fix: LiteLLM test failures after merge (thanks @daniel-lxs!)
-- Improve UX: Focus textbox and add newlines after adding to context (thanks @mrubens!)
-
 ## [3.29.2] - 2025-10-27
 
 - Add support for LongCat-Flash-Thinking-FP8 models in Chutes AI provider (#8425 by @leakless21, PR by @roomote)

CHANGELOG.md

@@ -1,15 +1,6 @@
 "use client"
 
-import {
-	ArrowRight,
-	Blocks,
-	BookMarked,
-	ListChecks,
-	LucideIcon,
-	GitPullRequest,
-	Key,
-	MessageSquareCode,
-} from "lucide-react"
+import { ArrowRight, Blocks, BookMarked, ListChecks, LucideIcon } from "lucide-react"
 import Image from "next/image"
 
 import { Button } from "@/components/ui"
@@ -25,26 +16,6 @@ interface Feature {
 	logos?: string[]
 }
 
-const workflowSteps: Feature[] = [
-	{
-		icon: GitPullRequest,
-		title: "1. Connect Your Repository",
-		description: "Link your GitHub repository and configure which branches and pull requests should be reviewed.",
-	},
-	{
-		icon: Key,
-		title: "2. Add Your API Key",
-		description:
-			"Provide your AI provider API key and set your review preferences, custom rules, and quality standards.",
-	},
-	{
-		icon: MessageSquareCode,
-		title: "3. Get Review Comments",
-		description:
-			"Every pull request gets detailed GitHub comments in minutes from a Roo Code agent highlighting issues and suggesting improvements.",
-	},
-]
-
 const howItWorks: Feature[] = [
 	{
 		icon: Blocks,
@@ -97,7 +68,7 @@ export function ReviewerContent() {
 									<p>
 										Roo Code&apos;s PR Reviewer flips the script: you bring your own key and
 										leverage it to the max – to find real issues, increase code quality and keep
-										your pull request queue moving.
+										your PR queue moving.
 									</p>
 								</div>
 							</div>
@@ -138,38 +109,6 @@ export function ReviewerContent() {
 				</div>
 			</section>
 
-			{/* How It Works Section */}
-			<section className="relative overflow-hidden border-t border-border py-32">
-				<div className="container relative z-10 mx-auto px-4 sm:px-6 lg:px-8">
-					<div className="mx-auto mb-12 md:mb-24 max-w-5xl text-center">
-						<div>
-							<h2 className="text-4xl font-bold tracking-tight sm:text-5xl">How It Works</h2>
-						</div>
-					</div>
-
-					<div className="relative mx-auto md:max-w-[1200px]">
-						<ul className="grid grid-cols-1 place-items-center gap-6 md:grid-cols-3 lg:gap-8">
-							{workflowSteps.map((step, index) => {
-								const Icon = step.icon
-								return (
-									<li
-										key={index}
-										className="relative h-full border border-border rounded-2xl bg-background p-8 transition-all duration-300 hover:shadow-lg">
-										<Icon className="size-6 text-foreground/80" />
-										<h3 className="mb-3 mt-3 text-xl font-semibold text-foreground">
-											{step.title}
-										</h3>
-										<div className="leading-relaxed font-light text-muted-foreground">
-											{step.description}
-										</div>
-									</li>
-								)
-							})}
-						</ul>
-					</div>
-				</div>
-			</section>
-
 			<section className="relative overflow-hidden border-t border-border py-32">
 				<div className="container relative z-10 mx-auto px-4 sm:px-6 lg:px-8">
 					<div className="mx-auto mb-12 md:mb-24 max-w-5xl text-center">

apps/web-roo-code/src/app/reviewer/ReviewerContent.tsx

@@ -93,17 +93,6 @@ export const globalSettingsSchema = z.object({
 	autoCondenseContextPercent: z.number().optional(),
 	maxConcurrentFileReads: z.number().optional(),
 
-	/**
-	 * Whether to include current time in the environment details
-	 * @default true
-	 */
-	includeCurrentTime: z.boolean().optional(),
-	/**
-	 * Whether to include current cost in the environment details
-	 * @default true
-	 */
-	includeCurrentCost: z.boolean().optional(),
-
 	/**
 	 * Whether to include diagnostic messages (errors, warnings) in tool outputs
 	 * @default true
@@ -218,7 +207,6 @@ export const SECRET_STATE_KEYS = [
 	"doubaoApiKey",
 	"moonshotApiKey",
 	"mistralApiKey",
-	"minimaxApiKey",
 	"unboundApiKey",
 	"requestyApiKey",
 	"xaiApiKey",

packages/types/src/global-settings.ts

@@ -61,13 +61,10 @@ export const modelInfoSchema = z.object({
 	// Capability flag to indicate whether the model supports an output verbosity parameter
 	supportsVerbosity: z.boolean().optional(),
 	supportsReasoningBudget: z.boolean().optional(),
-	// Capability flag to indicate whether the model supports simple on/off binary reasoning
-	supportsReasoningBinary: z.boolean().optional(),
 	// Capability flag to indicate whether the model supports temperature parameter
 	supportsTemperature: z.boolean().optional(),
 	requiredReasoningBudget: z.boolean().optional(),
 	supportsReasoningEffort: z.boolean().optional(),
-	requiredReasoningEffort: z.boolean().optional(),
 	supportedParameters: z.array(modelParametersSchema).optional(),
 	inputPrice: z.number().optional(),
 	outputPrice: z.number().optional(),

packages/types/src/model.ts

@@ -22,9 +22,7 @@ import {
 	sambaNovaModels,
 	vertexModels,
 	vscodeLlmModels,
-	xaiModels,
 	internationalZAiModels,
-	minimaxModels,
 } from "./providers/index.js"
 
 /**
@@ -50,6 +48,7 @@ export const dynamicProviders = [
 	"unbound",
 	"glama",
 	"roo",
+	"xai",
 ] as const
 
 export type DynamicProvider = (typeof dynamicProviders)[number]
@@ -132,13 +131,11 @@ export const providerNames = [
 	"groq",
 	"mistral",
 	"moonshot",
-	"minimax",
 	"openai-native",
 	"qwen-code",
 	"roo",
 	"sambanova",
 	"vertex",
-	"xai",
 	"zai",
 ] as const
 
@@ -329,13 +326,6 @@ const moonshotSchema = apiModelIdProviderModelSchema.extend({
 	moonshotApiKey: z.string().optional(),
 })
 
-const minimaxSchema = apiModelIdProviderModelSchema.extend({
-	minimaxBaseUrl: z
-		.union([z.literal("https://api.minimax.io/v1"), z.literal("https://api.minimaxi.com/v1")])
-		.optional(),
-	minimaxApiKey: z.string().optional(),
-})
-
 const unboundSchema = baseProviderSettingsSchema.extend({
 	unboundApiKey: z.string().optional(),
 	unboundModelId: z.string().optional(),
@@ -355,6 +345,7 @@ const fakeAiSchema = baseProviderSettingsSchema.extend({
 
 const xaiSchema = apiModelIdProviderModelSchema.extend({
 	xaiApiKey: z.string().optional(),
+	xaiModelContextWindow: z.number().optional(),
 })
 
 const groqSchema = apiModelIdProviderModelSchema.extend({
@@ -444,7 +435,6 @@ export const providerSettingsSchemaDiscriminated = z.discriminatedUnion("apiProv
 	deepInfraSchema.merge(z.object({ apiProvider: z.literal("deepinfra") })),
 	doubaoSchema.merge(z.object({ apiProvider: z.literal("doubao") })),
 	moonshotSchema.merge(z.object({ apiProvider: z.literal("moonshot") })),
-	minimaxSchema.merge(z.object({ apiProvider: z.literal("minimax") })),
 	unboundSchema.merge(z.object({ apiProvider: z.literal("unbound") })),
 	requestySchema.merge(z.object({ apiProvider: z.literal("requesty") })),
 	humanRelaySchema.merge(z.object({ apiProvider: z.literal("human-relay") })),
@@ -486,7 +476,6 @@ export const providerSettingsSchema = z.object({
 	...deepInfraSchema.shape,
 	...doubaoSchema.shape,
 	...moonshotSchema.shape,
-	...minimaxSchema.shape,
 	...unboundSchema.shape,
 	...requestySchema.shape,
 	...humanRelaySchema.shape,
@@ -571,7 +560,6 @@ export const modelIdKeysByProvider: Record<TypicalProvider, ModelIdKey> = {
 	"gemini-cli": "apiModelId",
 	mistral: "apiModelId",
 	moonshot: "apiModelId",
-	minimax: "apiModelId",
 	deepseek: "apiModelId",
 	deepinfra: "deepInfraModelId",
 	doubao: "apiModelId",
@@ -688,11 +676,6 @@ export const MODELS_BY_PROVIDER: Record<
 		label: "Moonshot",
 		models: Object.keys(moonshotModels),
 	},
-	minimax: {
-		id: "minimax",
-		label: "MiniMax",
-		models: Object.keys(minimaxModels),
-	},
 	"openai-native": {
 		id: "openai-native",
 		label: "OpenAI",
@@ -715,7 +698,7 @@ export const MODELS_BY_PROVIDER: Record<
 		label: "VS Code LM API",
 		models: Object.keys(vscodeLlmModels),
 	},
-	xai: { id: "xai", label: "xAI (Grok)", models: Object.keys(xaiModels) },
+	xai: { id: "xai", label: "xAI (Grok)", models: [] },
 	zai: { id: "zai", label: "Zai", models: Object.keys(internationalZAiModels) },
 
 	// Dynamic providers; models pulled from remote APIs.

packages/types/src/provider-settings.ts

@@ -3,18 +3,9 @@ import type { ModelInfo } from "../model.js"
 // https://inference-docs.cerebras.ai/api-reference/chat-completions
 export type CerebrasModelId = keyof typeof cerebrasModels
 
-export const cerebrasDefaultModelId: CerebrasModelId = "gpt-oss-120b"
+export const cerebrasDefaultModelId: CerebrasModelId = "qwen-3-coder-480b-free"
 
 export const cerebrasModels = {
-	"zai-glm-4.6": {
-		maxTokens: 16_384,
-		contextWindow: 128000,
-		supportsImages: false,
-		supportsPromptCache: false,
-		inputPrice: 0,
-		outputPrice: 0,
-		description: "Highly intelligent general-purpose model with ~2000 tokens/s",
-	},
 	"qwen-3-coder-480b-free": {
 		maxTokens: 40000,
 		contextWindow: 64000,
@@ -23,7 +14,7 @@ export const cerebrasModels = {
 		inputPrice: 0,
 		outputPrice: 0,
 		description:
-			"[SOON TO BE DEPRECATED] SOTA coding model with ~2000 tokens/s ($0 free tier)\n\n• Use this if you don't have a Cerebras subscription\n• 64K context window\n• Rate limits: 150K TPM, 1M TPH/TPD, 10 RPM, 100 RPH/RPD\n\nUpgrade for higher limits: [https://cloud.cerebras.ai/?utm=roocode](https://cloud.cerebras.ai/?utm=roocode)",
+			"SOTA coding model with ~2000 tokens/s ($0 free tier)\n\n• Use this if you don't have a Cerebras subscription\n• 64K context window\n• Rate limits: 150K TPM, 1M TPH/TPD, 10 RPM, 100 RPH/RPD\n\nUpgrade for higher limits: [https://cloud.cerebras.ai/?utm=roocode](https://cloud.cerebras.ai/?utm=roocode)",
 	},
 	"qwen-3-coder-480b": {
 		maxTokens: 40000,
@@ -33,7 +24,7 @@ export const cerebrasModels = {
 		inputPrice: 0,
 		outputPrice: 0,
 		description:
-			"[SOON TO BE DEPRECATED] SOTA coding model with ~2000 tokens/s ($50/$250 paid tiers)\n\n• Use this if you have a Cerebras subscription\n• 131K context window with higher rate limits",
+			"SOTA coding model with ~2000 tokens/s ($50/$250 paid tiers)\n\n• Use this if you have a Cerebras subscription\n• 131K context window with higher rate limits",
 	},
 	"qwen-3-235b-a22b-instruct-2507": {
 		maxTokens: 64000,