diff --git a/packages/ai/.changes/res-1273-catalog-generation-invariants.md b/packages/ai/.changes/res-1273-catalog-generation-invariants.md
new file mode 100644
index 0000000000..2e8951900f
--- /dev/null
+++ b/packages/ai/.changes/res-1273-catalog-generation-invariants.md
@@ -0,0 +1,5 @@
+- Fixed GitHub Copilot Grok and MAI-Code models failing on chat completions by routing them through the Copilot Responses API.
+- Raised the Codex subscription (openai-codex) context window for the GPT-5.6 family from 272k to the full 1,050,000 measured on the API side.
+- Added low/high reasoning effort levels for Kimi K3 on effort-capable providers (OpenRouter, Prime Inference, Hugging Face, Fireworks, opencode, Kimi Coding, Vercel); the Moonshot and GitHub Copilot transports cannot send effort, so their rows now advertise no selectable levels instead of a no-op max.
+- Fixed thinkingmachines/Inkling-Small advertising a maxTokens above its context window.
+- Catalog regeneration now validates invariants (maxTokens vs context window, Copilot API classification, Codex context divergence, cross-provider thinking-level consistency) and fails instead of writing a violating catalog.
diff --git a/packages/ai/scripts/generate-models.ts b/packages/ai/scripts/generate-models.ts
index 5a942ab3a7..b4cc9f8948 100644
--- a/packages/ai/scripts/generate-models.ts
+++ b/packages/ai/scripts/generate-models.ts
@@ -4,6 +4,13 @@ import { writeFileSync } from "fs";
 import { dirname, join } from "path";
 import { fileURLToPath } from "url";
 import { getAnthropicCacheCosts } from "../src/cache-pricing.js";
+import { supportsAdaptiveThinking } from "../src/providers/anthropic.js";
+import { getCompat } from "../src/providers/openai-completions.js";
+import {
+	CODEX_SMALLER_WINDOW_VERIFIED,
+	copilotModelApi,
+	validateModelCatalog,
+} from "./validate-model-catalog.js";
 import { COPILOT_CLIENT_HEADERS } from "../src/copilot-client-version.js";
 import { getOpenRouterReasoningCapabilities } from "../src/openrouter-reasoning.js";
 import {
@@ -96,9 +103,9 @@ const DEEPSEEK_V4_THINKING_LEVEL_MAP = {
 const KIMI_K3_THINKING_LEVEL_MAP = {
 	off: null,
 	minimal: null,
-	low: null,
+	low: "low",
 	medium: null,
-	high: null,
+	high: "high",
 	xhigh: null,
 	max: "max",
 } as const;
@@ -1257,17 +1264,8 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
 				if (m.tool_call !== true) continue;
 				if (m.status === "deprecated") continue;
 
-				// Copilot proxies Claude via the Anthropic Messages API
-				const isCopilotClaude = modelId.startsWith("claude-");
-				// gpt-5/gpt-6 models require responses API, others use completions
-				const needsResponsesApi =
-					modelId.startsWith("gpt-5") || modelId.startsWith("gpt-6") || modelId.startsWith("oswe");
-
-				const api: Api = isCopilotClaude
-					? "anthropic-messages"
-					: needsResponsesApi
-						? "openai-responses"
-						: "openai-completions";
+				// Unclassified families still ship a row so validation reports them by name instead of silently misrouting.
+				const api: Api = copilotModelApi(modelId) ?? "openai-completions";
 
 				const anthropicCompat =
 					api === "anthropic-messages" ? getAnthropicMessagesCompat("github-copilot", modelId) : undefined;
@@ -1456,6 +1454,14 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
 			}
 		}
 
+		// models.dev rows occasionally carry an output limit above the context window; clamp to keep the pair coherent.
+		for (const model of models) {
+			if (model.maxTokens > model.contextWindow) {
+				console.log(`Clamping ${model.provider}/${model.id} maxTokens ${model.maxTokens} -> ${model.contextWindow}`);
+				model.maxTokens = model.contextWindow;
+			}
+		}
+
 		console.log(`Loaded ${models.length} tool-capable models from models.dev`);
 		return models;
 	} catch (error) {
@@ -2014,6 +2020,14 @@ async function generateModels() {
 			maxTokens: CODEX_MAX_TOKENS,
 		},
 	];
+	// The ChatGPT backend accepts the 1M+ API-side window (#1597) except upstream-verified smaller rows.
+	for (const codexModel of codexModels) {
+		if (CODEX_SMALLER_WINDOW_VERIFIED.has(codexModel.id)) continue;
+		const openaiTwin = allModels.find((m) => m.provider === "openai" && m.id === codexModel.id);
+		if (openaiTwin && openaiTwin.contextWindow >= 1_000_000) {
+			codexModel.contextWindow = openaiTwin.contextWindow;
+		}
+	}
 	allModels.push(...codexModels);
 
 	// Add missing Grok models
@@ -2259,6 +2273,30 @@ async function generateModels() {
 		applyThinkingLevelMetadata(model);
 	}
 
+	// Non-adaptive anthropic-messages rows think via budget tokens, where xhigh/max clamp to high.
+	for (const model of allModels) {
+		if (model.api !== "anthropic-messages" || !model.reasoning || !model.thinkingLevelMap) continue;
+		if (supportsAdaptiveThinking(model.id)) continue;
+		const phantom = ["xhigh", "max"].filter((level) => model.thinkingLevelMap?.[level] != null);
+		if (phantom.length === 0) continue;
+		console.log(`Nulling budget-clamped thinking levels on ${model.provider}/${model.id}: ${phantom.join(", ")}`);
+		model.thinkingLevelMap = { ...model.thinkingLevelMap, ...Object.fromEntries(phantom.map((level) => [level, null])) };
+	}
+
+	// Family maps can land on transports that never send reasoning effort; null them so the UI offers nothing the request drops.
+	for (const model of allModels) {
+		if (model.api !== "openai-completions" || !model.reasoning || !model.thinkingLevelMap) continue;
+		const compat = getCompat(model as Model<"openai-completions">);
+		if (compat.thinkingFormat !== "openai" || compat.supportsReasoningEffort) continue;
+		const selectable = Object.entries(model.thinkingLevelMap).filter(([, mapped]) => mapped !== null);
+		if (selectable.length === 0) continue;
+		console.log(
+			`Nulling unsendable thinking levels on ${model.provider}/${model.id}: ${selectable.map(([level]) => level).join(", ")}`,
+		);
+		// All levels explicitly null: absent keys read as supported at runtime.
+		model.thinkingLevelMap = { off: null, minimal: null, low: null, medium: null, high: null, xhigh: null, max: null };
+	}
+
 	// Group by provider and deduplicate by model ID
 	const providers: Record<string, Record<string, Model<Api>>> = {};
 	for (const model of allModels) {
@@ -2272,6 +2310,14 @@ async function generateModels() {
 		}
 	}
 
+	const violations = validateModelCatalog(providers);
+	if (violations.length > 0) {
+		for (const violation of violations) {
+			console.error(`Catalog validation: ${violation}`);
+		}
+		throw new Error(`Model catalog validation failed with ${violations.length} violation(s); not writing catalog`);
+	}
+
 	// Generate TypeScript file. JSON string literals prevent remote catalog
 	// text from becoming executable source code.
 	const output = renderModelsFile(providers);
@@ -2294,4 +2340,7 @@ async function generateModels() {
 }
 
 // Run the generator
-generateModels().catch(console.error);
+generateModels().catch((error) => {
+	console.error(error);
+	process.exitCode = 1;
+});
diff --git a/packages/ai/scripts/validate-model-catalog.ts b/packages/ai/scripts/validate-model-catalog.ts
new file mode 100644
index 0000000000..937deaea01
--- /dev/null
+++ b/packages/ai/scripts/validate-model-catalog.ts
@@ -0,0 +1,132 @@
+import { getSupportedThinkingLevels } from "../src/models.js";
+import { supportsAdaptiveThinking } from "../src/providers/anthropic.js";
+import { getCompat } from "../src/providers/openai-completions.js";
+import type { Api, Model } from "../src/types.js";
+
+/** The subset of a catalog row the invariants read; MODELS rows satisfy it. */
+export interface CatalogRowLike {
+	id: string;
+	api: string;
+	provider: string;
+	contextWindow: number;
+	maxTokens: number;
+	reasoning: boolean;
+	baseUrl?: string;
+	thinkingLevelMap?: Readonly<Record<string, string | null>>;
+	compat?: Readonly<Record<string, unknown>>;
+}
+
+export type CatalogLike = Readonly<Record<string, Readonly<Record<string, CatalogRowLike>>>>;
+
+/** Copilot serves each model family through exactly one endpoint; unclassified ids fail validation. */
+export function copilotModelApi(modelId: string): Api | undefined {
+	if (modelId.startsWith("claude-")) return "anthropic-messages";
+	// Responses-only on Copilot; /chat/completions rejects these families (upstream pi-mono #906).
+	if (
+		modelId.startsWith("gpt-5") ||
+		modelId.startsWith("gpt-6") ||
+		modelId.startsWith("oswe") ||
+		modelId.startsWith("grok-") ||
+		modelId.startsWith("mai-")
+	) {
+		return "openai-responses";
+	}
+	if (modelId.startsWith("gemini-") || modelId.startsWith("kimi-")) return "openai-completions";
+	return undefined;
+}
+
+// ChatGPT-backend window verified smaller than the API side (Codex CLI models.json, rust-v0.153.4).
+export const CODEX_SMALLER_WINDOW_VERIFIED = new Set(["gpt-6-astra"]);
+
+function familyKey(modelId: string): string {
+	const segments = modelId.split("/");
+	return segments[segments.length - 1].toLowerCase();
+}
+
+// Runtime-selectable levels via the UI's own function; compared within one transport, declared maps only, modulo "off".
+function selectableLevels(model: CatalogRowLike): string {
+	return getSupportedThinkingLevels(model as Model<Api>)
+		.filter((level) => level !== "off")
+		.join(",");
+}
+
+// Plain openai-format completions gate reasoning params on compat; other formats use the map as an enable toggle.
+function effortIsSendable(model: CatalogRowLike): boolean {
+	// Model requires baseUrl, so rows without one behave like the empty-string rows: provider-only detection.
+	const compat = getCompat({ ...model, baseUrl: model.baseUrl ?? "" } as Model<"openai-completions">);
+	return compat.thinkingFormat !== "openai" || compat.supportsReasoningEffort;
+}
+
+/** Generation-time catalog invariants; an empty return means the catalog is valid. */
+export function validateModelCatalog(catalog: CatalogLike): string[] {
+	const violations: string[] = [];
+
+	for (const [provider, models] of Object.entries(catalog)) {
+		for (const model of Object.values(models)) {
+			if (model.maxTokens > model.contextWindow) {
+				violations.push(
+					`${provider}/${model.id}: maxTokens ${model.maxTokens} exceeds contextWindow ${model.contextWindow}`,
+				);
+			}
+		}
+	}
+
+	for (const model of Object.values(catalog["github-copilot"] ?? {})) {
+		const expectedApi = copilotModelApi(model.id);
+		if (expectedApi === undefined) {
+			violations.push(
+				`github-copilot/${model.id}: unclassified model family; add it to copilotModelApi in validate-model-catalog.ts`,
+			);
+		} else if (model.api !== expectedApi) {
+			violations.push(`github-copilot/${model.id}: api ${model.api} does not match classification ${expectedApi}`);
+		}
+	}
+
+	for (const model of Object.values(catalog["openai-codex"] ?? {})) {
+		const openaiTwin = catalog.openai?.[model.id];
+		if (!openaiTwin || CODEX_SMALLER_WINDOW_VERIFIED.has(model.id)) continue;
+		const ratio = openaiTwin.contextWindow / model.contextWindow;
+		if (ratio > 2 || ratio < 0.5) {
+			violations.push(
+				`openai-codex/${model.id}: contextWindow ${model.contextWindow} diverges more than 2x from openai/${model.id} (${openaiTwin.contextWindow})`,
+			);
+		}
+	}
+
+	const familyLevels = new Map<string, Map<string, string>>();
+	for (const [provider, models] of Object.entries(catalog)) {
+		for (const model of Object.values(models)) {
+			if (!model.thinkingLevelMap || !model.reasoning) continue;
+			if (model.api === "anthropic-messages" && !supportsAdaptiveThinking(model.id)) {
+				const clamped = ["xhigh", "max"].filter((level) => model.thinkingLevelMap?.[level] != null);
+				if (clamped.length > 0) {
+					violations.push(
+						`${provider}/${model.id}: thinkingLevelMap offers [${clamped.join(",")}] but the budget path serializes them as high`,
+					);
+				}
+			}
+			if (model.api === "openai-completions" && !effortIsSendable(model)) {
+				const levels = selectableLevels(model);
+				if (levels.length > 0) {
+					violations.push(
+						`${provider}/${model.id}: thinkingLevelMap offers [${levels}] but the transport cannot send reasoning effort`,
+					);
+				}
+				continue;
+			}
+			const key = `${familyKey(model.id)} [${model.api}]`;
+			const seen = familyLevels.get(key) ?? new Map<string, string>();
+			seen.set(`${provider}/${model.id}`, selectableLevels(model));
+			familyLevels.set(key, seen);
+		}
+	}
+	for (const [key, seen] of familyLevels) {
+		const distinct = new Set(seen.values());
+		if (distinct.size > 1) {
+			const detail = [...seen.entries()].map(([row, levels]) => `${row}=[${levels}]`).join(", ");
+			violations.push(`${key}: selectable thinking levels disagree across providers: ${detail}`);
+		}
+	}
+
+	return violations;
+}
diff --git a/packages/ai/src/models.generated.ts b/packages/ai/src/models.generated.ts
index ff0f332401..54bf0a61bc 100644
--- a/packages/ai/src/models.generated.ts
+++ b/packages/ai/src/models.generated.ts
@@ -886,6 +886,23 @@ export const MODELS = {
 			contextWindow: 1050000,
 			maxTokens: 128000,
 		} satisfies Model<"bedrock-converse-stream">,
+		"global.xai.grok-4.6": {
+			id: "global.xai.grok-4.6",
+			name: "Grok 4.6 (Global)",
+			api: "bedrock-converse-stream",
+			provider: "amazon-bedrock",
+			baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
+			reasoning: true,
+			input: ["text", "image"],
+			cost: {
+				input: 2,
+				output: 6,
+				cacheRead: 0.5,
+				cacheWrite: 0,
+			},
+			contextWindow: 500000,
+			maxTokens: 500000,
+		} satisfies Model<"bedrock-converse-stream">,
 		"google.gemma-3-27b-it": {
 			id: "google.gemma-3-27b-it",
 			name: "Google Gemma 3 27B Instruct",
@@ -2009,6 +2026,23 @@ export const MODELS = {
 			contextWindow: 3500000,
 			maxTokens: 16384,
 		} satisfies Model<"bedrock-converse-stream">,
+		"us.xai.grok-4.6": {
+			id: "us.xai.grok-4.6",
+			name: "Grok 4.6 (US)",
+			api: "bedrock-converse-stream",
+			provider: "amazon-bedrock",
+			baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
+			reasoning: true,
+			input: ["text", "image"],
+			cost: {
+				input: 2.2,
+				output: 6.6,
+				cacheRead: 0.55,
+				cacheWrite: 0,
+			},
+			contextWindow: 500000,
+			maxTokens: 500000,
+		} satisfies Model<"bedrock-converse-stream">,
 		"writer.palmyra-x4-v1:0": {
 			id: "writer.palmyra-x4-v1:0",
 			name: "Palmyra X4",
@@ -3119,39 +3153,39 @@ export const MODELS = {
 		} satisfies Model<"azure-openai-responses">,
 	},
 	"cerebras": {
-		"gemma-4-31b": {
-			id: "gemma-4-31b",
-			name: "Gemma 4 31B IT",
+		"gpt-oss-120b": {
+			id: "gpt-oss-120b",
+			name: "GPT OSS 120B",
 			api: "openai-completions",
 			provider: "cerebras",
 			baseUrl: "https://api.cerebras.ai/v1",
 			reasoning: true,
-			input: ["text", "image"],
+			input: ["text"],
 			cost: {
-				input: 0.99,
-				output: 1.49,
+				input: 0.35,
+				output: 0.75,
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
 			contextWindow: 131072,
 			maxTokens: 40960,
 		} satisfies Model<"openai-completions">,
-		"gpt-oss-120b": {
-			id: "gpt-oss-120b",
-			name: "GPT OSS 120B",
+		"qwen-3.8-27b": {
+			id: "qwen-3.8-27b",
+			name: "Qwen3.8 27B",
 			api: "openai-completions",
 			provider: "cerebras",
 			baseUrl: "https://api.cerebras.ai/v1",
 			reasoning: true,
-			input: ["text"],
+			input: ["text", "image"],
 			cost: {
-				input: 0.35,
-				output: 0.75,
+				input: 0.99,
+				output: 1.49,
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 131072,
-			maxTokens: 40960,
+			contextWindow: 65536,
+			maxTokens: 32768,
 		} satisfies Model<"openai-completions">,
 	},
 	"cloudflare-ai-gateway": {
@@ -4267,7 +4301,7 @@ export const MODELS = {
 			provider: "fireworks",
 			baseUrl: "https://api.fireworks.ai/inference",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -4421,7 +4455,7 @@ export const MODELS = {
 			provider: "fireworks",
 			baseUrl: "https://api.fireworks.ai/inference",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 4.5,
@@ -4855,11 +4889,10 @@ export const MODELS = {
 		"grok-4.5": {
 			id: "grok-4.5",
 			name: "Grok 4.5",
-			api: "openai-completions",
+			api: "openai-responses",
 			provider: "github-copilot",
 			baseUrl: "https://api.individual.githubcopilot.com",
 			headers: {"User-Agent":"GitHubCopilotChat/0.48.1","Editor-Version":"vscode/1.136.1","Editor-Plugin-Version":"copilot-chat/0.48.1","Copilot-Integration-Id":"vscode-chat"},
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
 			reasoning: true,
 			input: ["text", "image"],
 			cost: {
@@ -4870,15 +4903,14 @@ export const MODELS = {
 			},
 			contextWindow: 500000,
 			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
+		} satisfies Model<"openai-responses">,
 		"grok-4.6": {
 			id: "grok-4.6",
 			name: "Grok 4.6",
-			api: "openai-completions",
+			api: "openai-responses",
 			provider: "github-copilot",
 			baseUrl: "https://api.individual.githubcopilot.com",
 			headers: {"User-Agent":"GitHubCopilotChat/0.48.1","Editor-Version":"vscode/1.136.1","Editor-Plugin-Version":"copilot-chat/0.48.1","Copilot-Integration-Id":"vscode-chat"},
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
 			reasoning: true,
 			input: ["text", "image"],
 			cost: {
@@ -4889,7 +4921,7 @@ export const MODELS = {
 			},
 			contextWindow: 500000,
 			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
+		} satisfies Model<"openai-responses">,
 		"kimi-k2.7-code": {
 			id: "kimi-k2.7-code",
 			name: "Kimi K2.7 Code",
@@ -4918,7 +4950,7 @@ export const MODELS = {
 			headers: {"User-Agent":"GitHubCopilotChat/0.48.1","Editor-Version":"vscode/1.136.1","Editor-Plugin-Version":"copilot-chat/0.48.1","Copilot-Integration-Id":"vscode-chat"},
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -4932,11 +4964,10 @@ export const MODELS = {
 		"mai-code-1-flash-picker": {
 			id: "mai-code-1-flash-picker",
 			name: "MAI-Code-1-Flash",
-			api: "openai-completions",
+			api: "openai-responses",
 			provider: "github-copilot",
 			baseUrl: "https://api.individual.githubcopilot.com",
 			headers: {"User-Agent":"GitHubCopilotChat/0.48.1","Editor-Version":"vscode/1.136.1","Editor-Plugin-Version":"copilot-chat/0.48.1","Copilot-Integration-Id":"vscode-chat"},
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
 			reasoning: true,
 			input: ["text"],
 			cost: {
@@ -4947,15 +4978,14 @@ export const MODELS = {
 			},
 			contextWindow: 256000,
 			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
+		} satisfies Model<"openai-responses">,
 		"mai-code-1.1-flash": {
 			id: "mai-code-1.1-flash",
 			name: "MAI-Code-1.1-Flash",
-			api: "openai-completions",
+			api: "openai-responses",
 			provider: "github-copilot",
 			baseUrl: "https://api.individual.githubcopilot.com",
 			headers: {"User-Agent":"GitHubCopilotChat/0.48.1","Editor-Version":"vscode/1.136.1","Editor-Plugin-Version":"copilot-chat/0.48.1","Copilot-Integration-Id":"vscode-chat"},
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false},
 			reasoning: true,
 			input: ["text", "image"],
 			cost: {
@@ -4966,7 +4996,7 @@ export const MODELS = {
 			},
 			contextWindow: 256000,
 			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
+		} satisfies Model<"openai-responses">,
 	},
 	"google": {
 		"gemini-2.5-flash": {
@@ -6546,7 +6576,7 @@ export const MODELS = {
 			baseUrl: "https://router.huggingface.co/v1",
 			compat: {"supportsDeveloperRole":false},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":"max"},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -6681,7 +6711,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 524288,
-			maxTokens: 1048576,
+			maxTokens: 524288,
 		} satisfies Model<"openai-completions">,
 		"zai-org/GLM-4.5": {
 			id: "zai-org/GLM-4.5",
@@ -6909,7 +6939,7 @@ export const MODELS = {
 			baseUrl: "https://api.kimi.com/coding",
 			headers: {"User-Agent":"KimiCLI/1.5"},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0,
@@ -6928,7 +6958,7 @@ export const MODELS = {
 			baseUrl: "https://api.kimi.com/coding",
 			headers: {"User-Agent":"KimiCLI/1.5"},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0,
@@ -7595,120 +7625,87 @@ export const MODELS = {
 		} satisfies Model<"mistral-conversations">,
 	},
 	"moonshotai": {
-		"kimi-k2-0711-preview": {
-			id: "kimi-k2-0711-preview",
-			name: "Kimi K2 0711",
-			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
-				cacheWrite: 0,
-			},
-			contextWindow: 131072,
-			maxTokens: 16384,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-0905-preview": {
-			id: "kimi-k2-0905-preview",
-			name: "Kimi K2 0905",
+		"kimi-k2.6": {
+			id: "kimi-k2.6",
+			name: "Kimi K2.6",
 			api: "openai-completions",
 			provider: "moonshotai",
 			baseUrl: "https://api.moonshot.ai/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: false,
-			input: ["text"],
+			reasoning: true,
+			input: ["text", "image"],
 			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
+				input: 0.95,
+				output: 4,
+				cacheRead: 0.16,
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
 			maxTokens: 262144,
 		} satisfies Model<"openai-completions">,
-		"kimi-k2-thinking": {
-			id: "kimi-k2-thinking",
-			name: "Kimi K2 Thinking",
+		"kimi-k2.7-code": {
+			id: "kimi-k2.7-code",
+			name: "Kimi K2.7 Code",
 			api: "openai-completions",
 			provider: "moonshotai",
 			baseUrl: "https://api.moonshot.ai/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
-			input: ["text"],
+			input: ["text", "image"],
 			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
+				input: 0.95,
+				output: 4,
+				cacheRead: 0.19,
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
 			maxTokens: 262144,
 		} satisfies Model<"openai-completions">,
-		"kimi-k2-thinking-turbo": {
-			id: "kimi-k2-thinking-turbo",
-			name: "Kimi K2 Thinking Turbo",
+		"kimi-k2.7-code-highspeed": {
+			id: "kimi-k2.7-code-highspeed",
+			name: "Kimi K2.7 Code HighSpeed",
 			api: "openai-completions",
 			provider: "moonshotai",
 			baseUrl: "https://api.moonshot.ai/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
-			input: ["text"],
+			input: ["text", "image"],
 			cost: {
-				input: 1.15,
+				input: 1.9,
 				output: 8,
-				cacheRead: 0.15,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-turbo-preview": {
-			id: "kimi-k2-turbo-preview",
-			name: "Kimi K2 Turbo",
-			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 2.4,
-				output: 10,
-				cacheRead: 0.6,
+				cacheRead: 0.38,
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
 			maxTokens: 262144,
 		} satisfies Model<"openai-completions">,
-		"kimi-k2.5": {
-			id: "kimi-k2.5",
-			name: "Kimi K2.5",
+		"kimi-k3": {
+			id: "kimi-k3",
+			name: "Kimi K3",
 			api: "openai-completions",
 			provider: "moonshotai",
 			baseUrl: "https://api.moonshot.ai/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
-				input: 0.6,
-				output: 3,
-				cacheRead: 0.1,
+				input: 3,
+				output: 15,
+				cacheRead: 0.3,
 				cacheWrite: 0,
 			},
-			contextWindow: 262144,
-			maxTokens: 262144,
+			contextWindow: 1048576,
+			maxTokens: 131072,
 		} satisfies Model<"openai-completions">,
+	},
+	"moonshotai-cn": {
 		"kimi-k2.6": {
 			id: "kimi-k2.6",
 			name: "Kimi K2.6",
 			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
+			provider: "moonshotai-cn",
+			baseUrl: "https://api.moonshot.cn/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
 			input: ["text", "image"],
@@ -7725,8 +7722,8 @@ export const MODELS = {
 			id: "kimi-k2.7-code",
 			name: "Kimi K2.7 Code",
 			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
+			provider: "moonshotai-cn",
+			baseUrl: "https://api.moonshot.cn/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
 			input: ["text", "image"],
@@ -7743,8 +7740,8 @@ export const MODELS = {
 			id: "kimi-k2.7-code-highspeed",
 			name: "Kimi K2.7 Code HighSpeed",
 			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
+			provider: "moonshotai-cn",
+			baseUrl: "https://api.moonshot.cn/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
 			input: ["text", "image"],
@@ -7761,11 +7758,11 @@ export const MODELS = {
 			id: "kimi-k3",
 			name: "Kimi K3",
 			api: "openai-completions",
-			provider: "moonshotai",
-			baseUrl: "https://api.moonshot.ai/v1",
+			provider: "moonshotai-cn",
+			baseUrl: "https://api.moonshot.cn/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -7777,236 +7774,53 @@ export const MODELS = {
 			maxTokens: 131072,
 		} satisfies Model<"openai-completions">,
 	},
-	"moonshotai-cn": {
-		"kimi-k2-0711-preview": {
-			id: "kimi-k2-0711-preview",
-			name: "Kimi K2 0711",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
+	"openai": {
+		"gpt-4": {
+			id: "gpt-4",
+			name: "GPT-4",
+			api: "openai-responses",
+			provider: "openai",
+			baseUrl: "https://api.openai.com/v1",
 			reasoning: false,
 			input: ["text"],
 			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
+				input: 30,
+				output: 60,
+				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 131072,
-			maxTokens: 16384,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-0905-preview": {
-			id: "kimi-k2-0905-preview",
-			name: "Kimi K2 0905",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			contextWindow: 8192,
+			maxTokens: 8192,
+		} satisfies Model<"openai-responses">,
+		"gpt-4-turbo": {
+			id: "gpt-4-turbo",
+			name: "GPT-4 Turbo",
+			api: "openai-responses",
+			provider: "openai",
+			baseUrl: "https://api.openai.com/v1",
 			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-thinking": {
-			id: "kimi-k2-thinking",
-			name: "Kimi K2 Thinking",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text"],
+			input: ["text", "image"],
 			cost: {
-				input: 0.6,
-				output: 2.5,
-				cacheRead: 0.15,
+				input: 10,
+				output: 30,
+				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-thinking-turbo": {
-			id: "kimi-k2-thinking-turbo",
-			name: "Kimi K2 Thinking Turbo",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text"],
+			contextWindow: 128000,
+			maxTokens: 4096,
+		} satisfies Model<"openai-responses">,
+		"gpt-4.1": {
+			id: "gpt-4.1",
+			name: "GPT-4.1",
+			api: "openai-responses",
+			provider: "openai",
+			baseUrl: "https://api.openai.com/v1",
+			reasoning: false,
+			input: ["text", "image"],
 			cost: {
-				input: 1.15,
+				input: 2,
 				output: 8,
-				cacheRead: 0.15,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2-turbo-preview": {
-			id: "kimi-k2-turbo-preview",
-			name: "Kimi K2 Turbo",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 2.4,
-				output: 10,
-				cacheRead: 0.6,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2.5": {
-			id: "kimi-k2.5",
-			name: "Kimi K2.5",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text", "image"],
-			cost: {
-				input: 0.6,
-				output: 3,
-				cacheRead: 0.1,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2.6": {
-			id: "kimi-k2.6",
-			name: "Kimi K2.6",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text", "image"],
-			cost: {
-				input: 0.95,
-				output: 4,
-				cacheRead: 0.16,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2.7-code": {
-			id: "kimi-k2.7-code",
-			name: "Kimi K2.7 Code",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text", "image"],
-			cost: {
-				input: 0.95,
-				output: 4,
-				cacheRead: 0.19,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k2.7-code-highspeed": {
-			id: "kimi-k2.7-code-highspeed",
-			name: "Kimi K2.7 Code HighSpeed",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			input: ["text", "image"],
-			cost: {
-				input: 1.9,
-				output: 8,
-				cacheRead: 0.38,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 262144,
-		} satisfies Model<"openai-completions">,
-		"kimi-k3": {
-			id: "kimi-k3",
-			name: "Kimi K3",
-			api: "openai-completions",
-			provider: "moonshotai-cn",
-			baseUrl: "https://api.moonshot.cn/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
-			input: ["text", "image"],
-			cost: {
-				input: 3,
-				output: 15,
-				cacheRead: 0.3,
-				cacheWrite: 0,
-			},
-			contextWindow: 1048576,
-			maxTokens: 131072,
-		} satisfies Model<"openai-completions">,
-	},
-	"openai": {
-		"gpt-4": {
-			id: "gpt-4",
-			name: "GPT-4",
-			api: "openai-responses",
-			provider: "openai",
-			baseUrl: "https://api.openai.com/v1",
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 30,
-				output: 60,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 8192,
-			maxTokens: 8192,
-		} satisfies Model<"openai-responses">,
-		"gpt-4-turbo": {
-			id: "gpt-4-turbo",
-			name: "GPT-4 Turbo",
-			api: "openai-responses",
-			provider: "openai",
-			baseUrl: "https://api.openai.com/v1",
-			reasoning: false,
-			input: ["text", "image"],
-			cost: {
-				input: 10,
-				output: 30,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 128000,
-			maxTokens: 4096,
-		} satisfies Model<"openai-responses">,
-		"gpt-4.1": {
-			id: "gpt-4.1",
-			name: "GPT-4.1",
-			api: "openai-responses",
-			provider: "openai",
-			baseUrl: "https://api.openai.com/v1",
-			reasoning: false,
-			input: ["text", "image"],
-			cost: {
-				input: 2,
-				output: 8,
-				cacheRead: 0.5,
+				cacheRead: 0.5,
 				cacheWrite: 0,
 			},
 			contextWindow: 1047576,
@@ -8895,7 +8709,7 @@ export const MODELS = {
 				cacheRead: 0.1,
 				cacheWrite: 1.25,
 			},
-			contextWindow: 272000,
+			contextWindow: 1050000,
 			maxTokens: 128000,
 		} satisfies Model<"openai-codex-responses">,
 		"gpt-5.6-sol": {
@@ -8913,7 +8727,7 @@ export const MODELS = {
 				cacheRead: 0.5,
 				cacheWrite: 6.25,
 			},
-			contextWindow: 272000,
+			contextWindow: 1050000,
 			maxTokens: 128000,
 		} satisfies Model<"openai-codex-responses">,
 		"gpt-5.6-terra": {
@@ -8931,7 +8745,7 @@ export const MODELS = {
 				cacheRead: 0.25,
 				cacheWrite: 3.125,
 			},
-			contextWindow: 272000,
+			contextWindow: 1050000,
 			maxTokens: 128000,
 		} satisfies Model<"openai-codex-responses">,
 		"gpt-6-astra": {
@@ -9920,7 +9734,7 @@ export const MODELS = {
 			provider: "opencode",
 			baseUrl: "https://opencode.ai/zen/v1",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":"max"},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -10389,7 +10203,7 @@ export const MODELS = {
 			provider: "opencode-go",
 			baseUrl: "https://opencode.ai/zen/go/v1",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":"max"},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -11402,9 +11216,9 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"max","max":null},
 			input: ["text"],
 			cost: {
-				input: 0.08106,
-				output: 0.16212,
-				cacheRead: 0.016212,
+				input: 0.088606,
+				output: 0.177212,
+				cacheRead: 0.017721200000000003,
 				cacheWrite: 0,
 			},
 			contextWindow: 1048576,
@@ -11421,9 +11235,9 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"max","max":null},
 			input: ["text"],
 			cost: {
-				input: 0.049980000000000004,
-				output: 0.09996000000000001,
-				cacheRead: 0.009996000000000001,
+				input: 0.14,
+				output: 0.28,
+				cacheRead: 0.028,
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
@@ -11459,9 +11273,9 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"max","max":null},
 			input: ["text"],
 			cost: {
-				input: 0.7078319999999999,
-				output: 1.4156639999999998,
-				cacheRead: 0.058986000000000004,
+				input: 0.9552599999999999,
+				output: 1.9105199999999998,
+				cacheRead: 0.07960500000000001,
 				cacheWrite: 0,
 			},
 			contextWindow: 1048576,
@@ -11478,9 +11292,9 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"max","max":null},
 			input: ["text"],
 			cost: {
-				input: 1.12068,
-				output: 3.36204,
-				cacheRead: 0.037356,
+				input: 1.0494,
+				output: 3.1482,
+				cacheRead: 0.03498,
 				cacheWrite: 0,
 			},
 			contextWindow: 1048576,
@@ -14556,7 +14370,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
-			maxTokens: 32768,
+			maxTokens: 235929,
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-vl-235b-a22b-instruct": {
 			id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -14751,13 +14565,13 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
-				input: 0.55,
-				output: 3.5,
-				cacheRead: 0.22499999999999998,
+				input: 0.39,
+				output: 2.34,
+				cacheRead: 0,
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
-			maxTokens: 235929,
+			maxTokens: 65536,
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.5-9b": {
 			id: "qwen/qwen3.5-9b",
@@ -15563,13 +15377,13 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text"],
 			cost: {
-				input: 0.55,
-				output: 2.2,
-				cacheRead: 0.11,
+				input: 0.43,
+				output: 1.75,
+				cacheRead: 0.08,
 				cacheWrite: 0,
 			},
 			contextWindow: 204800,
-			maxTokens: 131072,
+			maxTokens: 16384,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-4.6v": {
 			id: "z-ai/glm-4.6v",
@@ -15703,24 +15517,6 @@ export const MODELS = {
 			contextWindow: 1048576,
 			maxTokens: 131072,
 		} satisfies Model<"openai-completions">,
-		"z-ai/glm-5.2:free": {
-			id: "z-ai/glm-5.2:free",
-			name: "Z.ai: GLM 5.2 (free)",
-			api: "openai-completions",
-			provider: "openrouter",
-			baseUrl: "https://openrouter.ai/api/v1",
-			reasoning: true,
-			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"xhigh","max":null},
-			input: ["text"],
-			cost: {
-				input: 0,
-				output: 0,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 256000,
-			maxTokens: 230400,
-		} satisfies Model<"openai-completions">,
 		"z-ai/glm-5.3": {
 			id: "z-ai/glm-5.3",
 			name: "Z.ai: GLM 5.3",
@@ -15733,11 +15529,11 @@ export const MODELS = {
 			cost: {
 				input: 1.4,
 				output: 4.4,
-				cacheRead: 0.14,
+				cacheRead: 0.26,
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
-			maxTokens: 262144,
+			maxTokens: 943718,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-5.3-flash": {
 			id: "z-ai/glm-5.3-flash",
@@ -15860,13 +15656,13 @@ export const MODELS = {
 			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":"max","max":null},
 			input: ["text"],
 			cost: {
-				input: 0.045,
-				output: 0.09,
-				cacheRead: 0.009,
+				input: 0.049999999999999996,
+				output: 0.16,
+				cacheRead: 0.013000000000000001,
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
-			maxTokens: 943718,
+			maxTokens: 393216,
 		} satisfies Model<"openai-completions">,
 		"~google/gemini-flash-latest": {
 			id: "~google/gemini-flash-latest",
@@ -15986,13 +15782,13 @@ export const MODELS = {
 			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":"max"},
 			input: ["text", "image"],
 			cost: {
-				input: 0.075,
-				output: 0.25,
-				cacheRead: 0.015,
+				input: 0.07125000000000001,
+				output: 0.2375,
+				cacheRead: 0.01425,
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
-			maxTokens: 943718,
+			maxTokens: 131072,
 		} satisfies Model<"openai-completions">,
 		"~z-ai/glm-latest": {
 			id: "~z-ai/glm-latest",
@@ -16004,58 +15800,225 @@ export const MODELS = {
 			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":"max"},
 			input: ["text"],
 			cost: {
-				input: 1.17,
-				output: 3.9600000000000004,
-				cacheRead: 0.234,
+				input: 1.12,
+				output: 3.52,
+				cacheRead: 0.20800000000000002,
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
-			maxTokens: 235929,
+			maxTokens: 943718,
 		} satisfies Model<"openai-completions">,
 	},
 	"prime-inference": {
-		"anthropic/claude-fable-5": {
-			id: "anthropic/claude-fable-5",
-			name: "Claude Fable 5",
+		"Qwen/Qwen3-235B-A22B-Instruct-2507": {
+			id: "Qwen/Qwen3-235B-A22B-Instruct-2507",
+			name: "Qwen3 235B A22B Instruct 2507",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
 			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
-			input: ["text", "image"],
+			reasoning: false,
+			input: ["text"],
 			cost: {
-				input: 10,
-				output: 50,
-				cacheRead: 1,
-				cacheWrite: 12.5,
+				input: 0.25,
+				output: 1,
+				cacheRead: 0,
+				cacheWrite: 0,
 			},
-			contextWindow: 1000000,
-			maxTokens: 128000,
-			featured: true,
+			contextWindow: 262144,
+			maxTokens: 16384,
 		} satisfies Model<"openai-completions">,
-		"anthropic/claude-fable-5.1": {
-			id: "anthropic/claude-fable-5.1",
-			name: "Claude Fable 5.1",
+		"Qwen/Qwen3-235B-A22B-Thinking-2507": {
+			id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
+			name: "Qwen3 235B A22B Thinking 2507",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false,"thinkingFormat":"openrouter"},
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
-			input: ["text", "image"],
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
+			input: ["text"],
 			cost: {
-				input: 10,
-				output: 50,
-				cacheRead: 1,
-				cacheWrite: 12.5,
+				input: 0.3,
+				output: 3,
+				cacheRead: 0,
+				cacheWrite: 0,
 			},
-			contextWindow: 1000000,
-			maxTokens: 128000,
+			contextWindow: 131072,
+			maxTokens: 117964,
 		} satisfies Model<"openai-completions">,
-		"anthropic/claude-haiku-4.5": {
-			id: "anthropic/claude-haiku-4.5",
-			name: "Claude Haiku 4.5",
+		"Qwen/Qwen3-30B-A3B-Thinking-2507": {
+			id: "Qwen/Qwen3-30B-A3B-Thinking-2507",
+			name: "Qwen3 30B A3B Thinking 2507",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false,"thinkingFormat":"openrouter"},
+			reasoning: true,
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
+			input: ["text"],
+			cost: {
+				input: 0.2,
+				output: 2.4,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 81920,
+			maxTokens: 32768,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3-VL-30B-A3B-Thinking": {
+			id: "Qwen/Qwen3-VL-30B-A3B-Thinking",
+			name: "Qwen3 VL 30B A3B Thinking",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false,"thinkingFormat":"openrouter"},
+			reasoning: true,
+			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
+			input: ["text", "image"],
+			cost: {
+				input: 0.2,
+				output: 2.4,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 262144,
+			maxTokens: 32768,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3.5-0.8B": {
+			id: "Qwen/Qwen3.5-0.8B",
+			name: "QWEN3.5 0.8B",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: false,
+			input: ["text"],
+			cost: {
+				input: 0.04,
+				output: 0.08,
+				cacheRead: 0.036,
+				cacheWrite: 0.04,
+			},
+			contextWindow: 128000,
+			maxTokens: 8192,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3.5-122B-A10B": {
+			id: "Qwen/Qwen3.5-122B-A10B",
+			name: "QWEN3.5 122B A10B",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false,"thinkingFormat":"openrouter"},
+			reasoning: true,
+			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
+			input: ["text", "image"],
+			cost: {
+				input: 0.3,
+				output: 0.9,
+				cacheRead: 0.27,
+				cacheWrite: 0.3,
+			},
+			contextWindow: 262144,
+			maxTokens: 81920,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3.5-2B": {
+			id: "Qwen/Qwen3.5-2B",
+			name: "QWEN3.5 2B",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: false,
+			input: ["text"],
+			cost: {
+				input: 0.06,
+				output: 0.18,
+				cacheRead: 0.054,
+				cacheWrite: 0.06,
+			},
+			contextWindow: 128000,
+			maxTokens: 8192,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3.5-4B": {
+			id: "Qwen/Qwen3.5-4B",
+			name: "QWEN3.5 4B",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: false,
+			input: ["text"],
+			cost: {
+				input: 0.1,
+				output: 0.3,
+				cacheRead: 0.09,
+				cacheWrite: 0.1,
+			},
+			contextWindow: 128000,
+			maxTokens: 8192,
+		} satisfies Model<"openai-completions">,
+		"Qwen/Qwen3.5-9B": {
+			id: "Qwen/Qwen3.5-9B",
+			name: "QWEN3.5 9B",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":false,"maxTokensField":"max_tokens","supportsStrictMode":false,"thinkingFormat":"openrouter"},
+			reasoning: true,
+			thinkingLevelMap: {"minimal":null,"low":null,"medium":null,"high":"high","xhigh":null,"max":null},
+			input: ["text", "image"],
+			cost: {
+				input: 0.18,
+				output: 0.54,
+				cacheRead: 0.162,
+				cacheWrite: 0.18,
+			},
+			contextWindow: 262144,
+			maxTokens: 235929,
+		} satisfies Model<"openai-completions">,
+		"anthropic/claude-fable-5": {
+			id: "anthropic/claude-fable-5",
+			name: "Claude Fable 5",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: true,
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
+			input: ["text", "image"],
+			cost: {
+				input: 10,
+				output: 50,
+				cacheRead: 1,
+				cacheWrite: 12.5,
+			},
+			contextWindow: 1000000,
+			maxTokens: 128000,
+			featured: true,
+		} satisfies Model<"openai-completions">,
+		"anthropic/claude-fable-5.1": {
+			id: "anthropic/claude-fable-5.1",
+			name: "Claude Fable 5.1",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: true,
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
+			input: ["text", "image"],
+			cost: {
+				input: 10,
+				output: 50,
+				cacheRead: 1,
+				cacheWrite: 12.5,
+			},
+			contextWindow: 1000000,
+			maxTokens: 128000,
+		} satisfies Model<"openai-completions">,
+		"anthropic/claude-haiku-4.5": {
+			id: "anthropic/claude-haiku-4.5",
+			name: "Claude Haiku 4.5",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16206,7 +16169,7 @@ export const MODELS = {
 				cacheRead: 0.30000000000000004,
 				cacheWrite: 3.75,
 			},
-			contextWindow: 200000,
+			contextWindow: 1000000,
 			maxTokens: 64000,
 		} satisfies Model<"openai-completions">,
 		"anthropic/claude-sonnet-4.5": {
@@ -16225,7 +16188,7 @@ export const MODELS = {
 				cacheRead: 0.30000000000000004,
 				cacheWrite: 3.75,
 			},
-			contextWindow: 200000,
+			contextWindow: 1000000,
 			maxTokens: 64000,
 			featured: true,
 		} satisfies Model<"openai-completions">,
@@ -16271,7 +16234,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-chat": {
 			id: "deepseek/deepseek-chat",
-			name: "Deepseek Chat",
+			name: "DeepSeek V3",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16289,7 +16252,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-chat-v3-0324": {
 			id: "deepseek/deepseek-chat-v3-0324",
-			name: "Deepseek Chat V3 0324",
+			name: "DeepSeek V3 0324",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16307,7 +16270,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-chat-v3.1": {
 			id: "deepseek/deepseek-chat-v3.1",
-			name: "Deepseek Chat V3.1",
+			name: "DeepSeek V3.1",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16326,7 +16289,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v3.1-terminus": {
 			id: "deepseek/deepseek-v3.1-terminus",
-			name: "Deepseek V3.1 Terminus",
+			name: "DeepSeek V3.1 Terminus",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16345,7 +16308,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v3.2": {
 			id: "deepseek/deepseek-v3.2",
-			name: "Deepseek V3.2",
+			name: "DeepSeek V3.2",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16365,7 +16328,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v3.2-exp": {
 			id: "deepseek/deepseek-v3.2-exp",
-			name: "Deepseek V3.2 EXP",
+			name: "DeepSeek V3.2 Exp",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16384,7 +16347,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v4-flash": {
 			id: "deepseek/deepseek-v4-flash",
-			name: "Deepseek V4 Flash",
+			name: "DeepSeek V4 Flash 0423",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16404,7 +16367,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v4-flash-0731": {
 			id: "deepseek/deepseek-v4-flash-0731",
-			name: "Deepseek V4 Flash 0731",
+			name: "DeepSeek V4 Flash 0731",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16423,7 +16386,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"deepseek/deepseek-v4-pro": {
 			id: "deepseek/deepseek-v4-pro",
-			name: "Deepseek V4 PRO",
+			name: "DeepSeek V4 Pro 0423",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16481,7 +16444,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"google/gemini-2.5-pro": {
 			id: "google/gemini-2.5-pro",
-			name: "Gemini 2.5 PRO",
+			name: "Gemini 2.5 Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16519,7 +16482,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"google/gemini-3.1-pro-preview": {
 			id: "google/gemini-3.1-pro-preview",
-			name: "Gemini 3.1 PRO Preview",
+			name: "Gemini 3.1 Pro Preview",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16614,7 +16577,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"google/gemma-3-27b-it": {
 			id: "google/gemma-3-27b-it",
-			name: "Gemma 3 27B IT",
+			name: "Gemma 3 27B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16663,8 +16626,8 @@ export const MODELS = {
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 80000,
-			maxTokens: 80000,
+			contextWindow: 131072,
+			maxTokens: 117964,
 		} satisfies Model<"openai-completions">,
 		"meta-llama/llama-3.3-70b-instruct": {
 			id: "meta-llama/llama-3.3-70b-instruct",
@@ -16704,7 +16667,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"minimax/minimax-m2.5": {
 			id: "minimax/minimax-m2.5",
-			name: "Minimax M2.5",
+			name: "MiniMax M2.5",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16723,7 +16686,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"minimax/minimax-m2.7": {
 			id: "minimax/minimax-m2.7",
-			name: "Minimax M2.7",
+			name: "MiniMax M2.7",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16742,7 +16705,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"minimax/minimax-m3": {
 			id: "minimax/minimax-m3",
-			name: "Minimax M3",
+			name: "MiniMax M3",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16756,13 +16719,13 @@ export const MODELS = {
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 524288,
+			contextWindow: 1048576,
 			maxTokens: 512000,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"mistralai/mistral-large-2512": {
 			id: "mistralai/mistral-large-2512",
-			name: "Mistral Large 2512",
+			name: "Mistral Large 3 2512",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16798,7 +16761,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"mistralai/mistral-small-2603": {
 			id: "mistralai/mistral-small-2603",
-			name: "Mistral Small 2603",
+			name: "Mistral Small 4",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16817,7 +16780,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"mistralai/mixtral-8x22b-instruct": {
 			id: "mistralai/mixtral-8x22b-instruct",
-			name: "Mixtral 8X22B Instruct",
+			name: "Mixtral 8x22B Instruct",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16848,8 +16811,8 @@ export const MODELS = {
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 98304,
-			maxTokens: 98304,
+			contextWindow: 262144,
+			maxTokens: 100352,
 		} satisfies Model<"openai-completions">,
 		"moonshotai/kimi-k2.5": {
 			id: "moonshotai/kimi-k2.5",
@@ -16868,7 +16831,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
-			maxTokens: 65535,
+			maxTokens: 235929,
 		} satisfies Model<"openai-completions">,
 		"moonshotai/kimi-k2.6": {
 			id: "moonshotai/kimi-k2.6",
@@ -16887,7 +16850,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 262144,
-			maxTokens: 262144,
+			maxTokens: 235929,
 		} satisfies Model<"openai-completions">,
 		"moonshotai/kimi-k2.7-code": {
 			id: "moonshotai/kimi-k2.7-code",
@@ -16926,7 +16889,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 1048576,
-			maxTokens: 1048576,
+			maxTokens: 943718,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"nvidia/nemotron-3-nano-30b-a3b": {
@@ -16951,7 +16914,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"nvidia/nemotron-3-super-120b-a12b": {
 			id: "nvidia/nemotron-3-super-120b-a12b",
-			name: "Nemotron 3 Super 120B A12B",
+			name: "Nemotron 3 Super",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16965,13 +16928,13 @@ export const MODELS = {
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 262144,
-			maxTokens: 4096,
+			contextWindow: 1000000,
+			maxTokens: 16384,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-4.1": {
 			id: "openai/gpt-4.1",
-			name: "GPT 4.1",
+			name: "GPT-4.1",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -16989,7 +16952,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-4.1-mini": {
 			id: "openai/gpt-4.1-mini",
-			name: "GPT 4.1 Mini",
+			name: "GPT-4.1 Mini",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17007,7 +16970,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-4.1-nano": {
 			id: "openai/gpt-4.1-nano",
-			name: "GPT 4.1 Nano",
+			name: "GPT-4.1 Nano",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17025,7 +16988,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-4o": {
 			id: "openai/gpt-4o",
-			name: "GPT 4O",
+			name: "GPT-4o",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17043,7 +17006,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-4o-mini": {
 			id: "openai/gpt-4o-mini",
-			name: "GPT 4O Mini",
+			name: "GPT-4o-mini",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17061,7 +17024,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5": {
 			id: "openai/gpt-5",
-			name: "GPT 5",
+			name: "GPT-5",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17080,7 +17043,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5-mini": {
 			id: "openai/gpt-5-mini",
-			name: "GPT 5 Mini",
+			name: "GPT-5 Mini",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17099,7 +17062,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5-nano": {
 			id: "openai/gpt-5-nano",
-			name: "GPT 5 Nano",
+			name: "GPT-5 Nano",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17118,7 +17081,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.1": {
 			id: "openai/gpt-5.1",
-			name: "GPT 5.1",
+			name: "GPT-5.1",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17137,7 +17100,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.2": {
 			id: "openai/gpt-5.2",
-			name: "GPT 5.2",
+			name: "GPT-5.2",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17156,7 +17119,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.2-pro": {
 			id: "openai/gpt-5.2-pro",
-			name: "GPT 5.2 PRO",
+			name: "GPT-5.2 Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17175,7 +17138,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.3-codex": {
 			id: "openai/gpt-5.3-codex",
-			name: "GPT 5.3 Codex",
+			name: "GPT-5.3-Codex",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17195,7 +17158,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.4": {
 			id: "openai/gpt-5.4",
-			name: "GPT 5.4",
+			name: "GPT-5.4",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17215,7 +17178,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.4-mini": {
 			id: "openai/gpt-5.4-mini",
-			name: "GPT 5.4 Mini",
+			name: "GPT-5.4 Mini",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17235,7 +17198,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.4-nano": {
 			id: "openai/gpt-5.4-nano",
-			name: "GPT 5.4 Nano",
+			name: "GPT-5.4 Nano",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17254,7 +17217,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.4-pro": {
 			id: "openai/gpt-5.4-pro",
-			name: "GPT 5.4 PRO",
+			name: "GPT-5.4 Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17274,7 +17237,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.5": {
 			id: "openai/gpt-5.5",
-			name: "GPT 5.5",
+			name: "GPT-5.5",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17311,47 +17274,9 @@ export const MODELS = {
 			contextWindow: 128000,
 			maxTokens: 8192,
 		} satisfies Model<"openai-completions">,
-		"openai/gpt-5.6-luna": {
-			id: "openai/gpt-5.6-luna",
-			name: "GPT 5.6 Luna",
-			api: "openai-completions",
-			provider: "prime-inference",
-			baseUrl: "https://api.pinference.ai/api/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			thinkingLevelMap: {"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
-			input: ["text", "image"],
-			cost: {
-				input: 0.2,
-				output: 1.2,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 1050000,
-			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
-		"openai/gpt-5.6-luna-pro": {
-			id: "openai/gpt-5.6-luna-pro",
-			name: "GPT 5.6 Luna PRO",
-			api: "openai-completions",
-			provider: "prime-inference",
-			baseUrl: "https://api.pinference.ai/api/v1",
-			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
-			reasoning: true,
-			thinkingLevelMap: {"minimal":null,"low":"low","medium":"medium","high":"high","xhigh":"xhigh","max":"max"},
-			input: ["text", "image"],
-			cost: {
-				input: 1,
-				output: 6,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 1050000,
-			maxTokens: 128000,
-		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.6-sol": {
 			id: "openai/gpt-5.6-sol",
-			name: "GPT 5.6 SOL",
+			name: "GPT-5.6 Sol",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17370,7 +17295,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.6-sol-pro": {
 			id: "openai/gpt-5.6-sol-pro",
-			name: "GPT 5.6 SOL PRO",
+			name: "GPT-5.6 Sol Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17389,7 +17314,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.6-terra": {
 			id: "openai/gpt-5.6-terra",
-			name: "GPT 5.6 Terra",
+			name: "GPT-5.6 Terra",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17408,7 +17333,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-5.6-terra-pro": {
 			id: "openai/gpt-5.6-terra-pro",
-			name: "GPT 5.6 Terra PRO",
+			name: "GPT-5.6 Terra Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17427,7 +17352,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-6-astra": {
 			id: "openai/gpt-6-astra",
-			name: "GPT 6 Astra",
+			name: "GPT-6 Astra",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17446,7 +17371,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-oss-120b": {
 			id: "openai/gpt-oss-120b",
-			name: "GPT OSS 120B",
+			name: "gpt-oss-120b",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17465,7 +17390,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"openai/gpt-oss-20b": {
 			id: "openai/gpt-oss-20b",
-			name: "GPT OSS 20B",
+			name: "gpt-oss-20b",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17480,7 +17405,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 131072,
-			maxTokens: 131072,
+			maxTokens: 117964,
 		} satisfies Model<"openai-completions">,
 		"poolside/laguna-s-2.1": {
 			id: "poolside/laguna-s-2.1",
@@ -17522,7 +17447,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-235b-a22b-2507": {
 			id: "qwen/qwen3-235b-a22b-2507",
-			name: "QWEN3 235B A22B 2507",
+			name: "Qwen3 235B A22B Instruct 2507",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17540,7 +17465,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-30b-a3b-instruct-2507": {
 			id: "qwen/qwen3-30b-a3b-instruct-2507",
-			name: "QWEN3 30B A3B Instruct 2507",
+			name: "Qwen3 30B A3B Instruct 2507",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17559,7 +17484,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-8b": {
 			id: "qwen/qwen3-8b",
-			name: "QWEN3 8B",
+			name: "Qwen3 8B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17578,7 +17503,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-coder": {
 			id: "qwen/qwen3-coder",
-			name: "QWEN3 Coder",
+			name: "Qwen3 Coder 480B A35B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17596,7 +17521,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-coder-next": {
 			id: "qwen/qwen3-coder-next",
-			name: "QWEN3 Coder Next",
+			name: "Qwen3 Coder Next",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17615,7 +17540,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-max": {
 			id: "qwen/qwen3-max",
-			name: "QWEN3 MAX",
+			name: "Qwen3 Max",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17634,7 +17559,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-vl-235b-a22b-instruct": {
 			id: "qwen/qwen3-vl-235b-a22b-instruct",
-			name: "QWEN3 VL 235B A22B Instruct",
+			name: "Qwen3 VL 235B A22B Instruct",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17652,7 +17577,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-vl-235b-a22b-thinking": {
 			id: "qwen/qwen3-vl-235b-a22b-thinking",
-			name: "QWEN3 VL 235B A22B Thinking",
+			name: "Qwen3 VL 235B A22B Thinking",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17672,7 +17597,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-vl-30b-a3b-instruct": {
 			id: "qwen/qwen3-vl-30b-a3b-instruct",
-			name: "QWEN3 VL 30B A3B Instruct",
+			name: "Qwen3 VL 30B A3B Instruct",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17690,7 +17615,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3-vl-8b-instruct": {
 			id: "qwen/qwen3-vl-8b-instruct",
-			name: "QWEN3 VL 8B Instruct",
+			name: "Qwen3 VL 8B Instruct",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17708,7 +17633,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.5-35b-a3b": {
 			id: "qwen/qwen3.5-35b-a3b",
-			name: "QWEN3.5 35B A3B",
+			name: "Qwen3.5-35B-A3B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17727,7 +17652,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.5-397b-a17b": {
 			id: "qwen/qwen3.5-397b-a17b",
-			name: "QWEN3.5 397B A17B",
+			name: "Qwen3.5 397B A17B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17746,7 +17671,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.6-27b": {
 			id: "qwen/qwen3.6-27b",
-			name: "QWEN3.6 27B",
+			name: "Qwen3.6 27B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17765,7 +17690,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.6-35b-a3b": {
 			id: "qwen/qwen3.6-35b-a3b",
-			name: "QWEN3.6 35B A3B",
+			name: "Qwen3.6 35B A3B",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17784,7 +17709,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"qwen/qwen3.7-flash": {
 			id: "qwen/qwen3.7-flash",
-			name: "QWEN3.7 Flash",
+			name: "Qwen3.7 Flash",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17838,12 +17763,12 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 2000000,
-			maxTokens: 30000,
+			maxTokens: 1800000,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"x-ai/grok-4.20-multi-agent": {
 			id: "x-ai/grok-4.20-multi-agent",
-			name: "Grok 4.20 Multi Agent",
+			name: "Grok 4.20 Multi-Agent",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17858,7 +17783,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 2000000,
-			maxTokens: 30000,
+			maxTokens: 1800000,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"x-ai/grok-4.6": {
@@ -17882,7 +17807,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"xiaomi/mimo-v2.5": {
 			id: "xiaomi/mimo-v2.5",
-			name: "Mimo V2.5",
+			name: "MiMo-V2.5",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17901,7 +17826,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"xiaomi/mimo-v2.5-pro": {
 			id: "xiaomi/mimo-v2.5-pro",
-			name: "Mimo V2.5 PRO",
+			name: "MiMo-V2.5-Pro",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17939,7 +17864,7 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-4.5-air": {
 			id: "z-ai/glm-4.5-air",
-			name: "GLM 4.5 AIR",
+			name: "GLM 4.5 Air",
 			api: "openai-completions",
 			provider: "prime-inference",
 			baseUrl: "https://api.pinference.ai/api/v1",
@@ -17973,7 +17898,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 204800,
-			maxTokens: 131072,
+			maxTokens: 16384,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-4.7": {
 			id: "z-ai/glm-4.7",
@@ -18030,7 +17955,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 204800,
-			maxTokens: 131072,
+			maxTokens: 128000,
 			featured: true,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-5.1": {
@@ -18090,7 +18015,7 @@ export const MODELS = {
 				cacheWrite: 0,
 			},
 			contextWindow: 1310720,
-			maxTokens: 262144,
+			maxTokens: 943718,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-5.3-flash": {
 			id: "z-ai/glm-5.3-flash",
@@ -18111,6 +18036,24 @@ export const MODELS = {
 			contextWindow: 1310720,
 			maxTokens: 131072,
 		} satisfies Model<"openai-completions">,
+		"zai-org/GLM-4.7": {
+			id: "zai-org/GLM-4.7",
+			name: "GLM 4.7",
+			api: "openai-completions",
+			provider: "prime-inference",
+			baseUrl: "https://api.pinference.ai/api/v1",
+			compat: {"supportsStore":false,"supportsDeveloperRole":false,"supportsReasoningEffort":true,"maxTokensField":"max_tokens","supportsStrictMode":false},
+			reasoning: true,
+			input: ["text"],
+			cost: {
+				input: 0.6,
+				output: 2.2,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 204800,
+			maxTokens: 131072,
+		} satisfies Model<"openai-completions">,
 	},
 	"vercel-ai-gateway": {
 		"alibaba/qwen-3-14b": {
@@ -19942,23 +19885,6 @@ export const MODELS = {
 			contextWindow: 204800,
 			maxTokens: 131000,
 		} satisfies Model<"anthropic-messages">,
-		"minimax/minimax-m2.7-free": {
-			id: "minimax/minimax-m2.7-free",
-			name: "MiniMax M2.7 (Free)",
-			api: "anthropic-messages",
-			provider: "vercel-ai-gateway",
-			baseUrl: "https://ai-gateway.vercel.sh",
-			reasoning: true,
-			input: ["text"],
-			cost: {
-				input: 0,
-				output: 0,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 196608,
-			maxTokens: 196608,
-		} satisfies Model<"anthropic-messages">,
 		"minimax/minimax-m2.7-highspeed": {
 			id: "minimax/minimax-m2.7-highspeed",
 			name: "MiniMax M2.7 High Speed",
@@ -19993,23 +19919,6 @@ export const MODELS = {
 			contextWindow: 512000,
 			maxTokens: 512000,
 		} satisfies Model<"anthropic-messages">,
-		"minimax/minimax-m3-free": {
-			id: "minimax/minimax-m3-free",
-			name: "MiniMax M3 (Free)",
-			api: "anthropic-messages",
-			provider: "vercel-ai-gateway",
-			baseUrl: "https://ai-gateway.vercel.sh",
-			reasoning: true,
-			input: ["text", "image"],
-			cost: {
-				input: 0,
-				output: 0,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 1048576,
-			maxTokens: 1048576,
-		} satisfies Model<"anthropic-messages">,
 		"mistral/codestral": {
 			id: "mistral/codestral",
 			name: "Mistral Codestral",
@@ -20323,7 +20232,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3,
@@ -20341,7 +20250,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"off":null,"minimal":null,"low":null,"medium":null,"high":null,"xhigh":null,"max":"max"},
+			thinkingLevelMap: {"off":null,"minimal":null,"low":"low","medium":null,"high":"high","xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 4.5,
@@ -20869,7 +20778,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 1.75,
@@ -20887,7 +20796,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 1.75,
@@ -20905,7 +20814,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3.5,
@@ -20923,7 +20832,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 21,
@@ -20941,7 +20850,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 1.75,
@@ -20959,7 +20868,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 3.5,
@@ -20977,7 +20886,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 2.5,
@@ -20995,7 +20904,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 5,
@@ -21013,7 +20922,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0.75,
@@ -21031,7 +20940,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 1.5,
@@ -21049,7 +20958,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0.19999999999999998,
@@ -21067,7 +20976,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 30,
@@ -21085,7 +20994,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 5,
@@ -21103,7 +21012,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 12.5,
@@ -21121,7 +21030,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh"},
+			thinkingLevelMap: {"xhigh":null},
 			input: ["text", "image"],
 			cost: {
 				input: 30,
@@ -21139,7 +21048,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0.19999999999999998,
@@ -21157,7 +21066,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 0.39999999999999997,
@@ -21175,7 +21084,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 2,
@@ -21193,7 +21102,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 4,
@@ -21211,7 +21120,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 2,
@@ -21229,7 +21138,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"xhigh":"xhigh","minimal":null,"max":"max"},
+			thinkingLevelMap: {"xhigh":null,"minimal":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 4,
@@ -21247,7 +21156,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"minimal":null,"xhigh":"xhigh","max":"max"},
+			thinkingLevelMap: {"minimal":null,"xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 10,
@@ -21265,7 +21174,7 @@ export const MODELS = {
 			provider: "vercel-ai-gateway",
 			baseUrl: "https://ai-gateway.vercel.sh",
 			reasoning: true,
-			thinkingLevelMap: {"minimal":null,"xhigh":"xhigh","max":"max"},
+			thinkingLevelMap: {"minimal":null,"xhigh":null,"max":null},
 			input: ["text", "image"],
 			cost: {
 				input: 20,
diff --git a/packages/ai/src/providers/anthropic.ts b/packages/ai/src/providers/anthropic.ts
index 0be96d24ec..ac65e7ca5c 100644
--- a/packages/ai/src/providers/anthropic.ts
+++ b/packages/ai/src/providers/anthropic.ts
@@ -756,9 +756,10 @@ function isAlwaysOnAdaptiveThinkingModel(modelId: string): boolean {
 }
 
 /**
- * Check if a model supports adaptive thinking (Opus 4.6+, Sonnet 4.6)
+ * Check if a model supports adaptive thinking (Opus 4.6+, Sonnet 4.6). Exported
+ * for catalog generation, which must not re-derive the id list.
  */
-function supportsAdaptiveThinking(modelId: string): boolean {
+export function supportsAdaptiveThinking(modelId: string): boolean {
 	// Adaptive-thinking model IDs (with or without date suffix).
 	return (
 		modelId.includes("opus-4-6") ||
diff --git a/packages/ai/src/providers/openai-completions.ts b/packages/ai/src/providers/openai-completions.ts
index 78496a7593..a1b14868cc 100644
--- a/packages/ai/src/providers/openai-completions.ts
+++ b/packages/ai/src/providers/openai-completions.ts
@@ -1224,8 +1224,9 @@ function detectCompat(model: Model<"openai-completions">): ResolvedOpenAIComplet
 /**
  * Get resolved compatibility settings for a model.
  * Uses explicit model.compat if provided, otherwise auto-detects from provider/URL.
+ * Exported for catalog generation, which must not re-derive these rules.
  */
-function getCompat(model: Model<"openai-completions">): ResolvedOpenAICompletionsCompat {
+export function getCompat(model: Model<"openai-completions">): ResolvedOpenAICompletionsCompat {
 	const detected = detectCompat(model);
 	if (!model.compat) return detected;
 
diff --git a/packages/ai/test/model-catalog-validation.test.ts b/packages/ai/test/model-catalog-validation.test.ts
new file mode 100644
index 0000000000..d2d20c7568
--- /dev/null
+++ b/packages/ai/test/model-catalog-validation.test.ts
@@ -0,0 +1,113 @@
+import { describe, expect, it } from "vitest";
+import {
+	type CatalogLike,
+	type CatalogRowLike,
+	validateModelCatalog,
+} from "../scripts/validate-model-catalog.js";
+import { MODELS } from "../src/models.generated.js";
+
+function row(overrides: Partial<CatalogRowLike> & { id: string; provider: string }): CatalogRowLike {
+	return {
+		api: "openai-completions",
+		baseUrl: "https://example.com/v1",
+		reasoning: false,
+		contextWindow: 128000,
+		maxTokens: 8192,
+		...overrides,
+	};
+}
+
+function catalogOf(...models: CatalogRowLike[]): CatalogLike {
+	const catalog: Record<string, Record<string, CatalogRowLike>> = {};
+	for (const model of models) {
+		catalog[model.provider] ??= {};
+		catalog[model.provider][model.id] = model;
+	}
+	return catalog;
+}
+
+const NULL_MAP = { off: null, minimal: null, low: null, medium: null, high: null, xhigh: null, max: "max" };
+const RICH_MAP = { minimal: null, low: "low", medium: null, high: "high", xhigh: null, max: "max" };
+
+describe("model catalog validation", () => {
+	it("accepts the committed catalog", () => {
+		expect(validateModelCatalog(MODELS)).toEqual([]);
+	});
+
+	it.each<[string, CatalogLike, string[]]>([
+		[
+			"maxTokens above contextWindow",
+			catalogOf(row({ id: "swapped", provider: "huggingface", contextWindow: 1000, maxTokens: 2000 })),
+			["huggingface/swapped: maxTokens 2000 exceeds contextWindow 1000"],
+		],
+		[
+			"unclassified and misrouted GitHub Copilot models",
+			catalogOf(row({ id: "novel-model-x", provider: "github-copilot" }), row({ id: "grok-9", provider: "github-copilot" })),
+			[
+				"github-copilot/novel-model-x: unclassified model family; add it to copilotModelApi in validate-model-catalog.ts",
+				"github-copilot/grok-9: api openai-completions does not match classification openai-responses",
+			],
+		],
+		[
+			"codex contextWindow diverging more than 2x from the openai row",
+			catalogOf(
+				row({ id: "gpt-9", provider: "openai", api: "openai-responses", contextWindow: 1050000 }),
+				row({ id: "gpt-9", provider: "openai-codex", api: "openai-codex-responses", contextWindow: 272000 }),
+			),
+			["openai-codex/gpt-9: contextWindow 272000 diverges more than 2x from openai/gpt-9 (1050000)"],
+		],
+		[
+			"same-model rows whose runtime-selectable thinking levels disagree",
+			catalogOf(
+				row({ id: "vendor/model-y", provider: "openrouter", reasoning: true, thinkingLevelMap: RICH_MAP }),
+				row({ id: "model-y", provider: "huggingface", reasoning: true, thinkingLevelMap: NULL_MAP }),
+			),
+			[
+				"model-y [openai-completions]: selectable thinking levels disagree across providers: " +
+					"openrouter/vendor/model-y=[low,high,max], huggingface/model-y=[max]",
+			],
+		],
+		[
+			"selectable levels on a transport that cannot send reasoning effort, without crashing on a missing baseUrl",
+			catalogOf(
+				row({ id: "model-z", provider: "moonshotai", reasoning: true, thinkingLevelMap: NULL_MAP, baseUrl: undefined }),
+			),
+			["moonshotai/model-z: thinkingLevelMap offers [max] but the transport cannot send reasoning effort"],
+		],
+		[
+			"budget-clamped levels on a non-adaptive anthropic-messages row",
+			catalogOf(
+				row({
+					id: "kimi-x",
+					provider: "kimi-coding",
+					api: "anthropic-messages",
+					reasoning: true,
+					thinkingLevelMap: { off: null, low: "low", high: "high", max: "max" },
+				}),
+			),
+			["kimi-coding/kimi-x: thinkingLevelMap offers [max] but the budget path serializes them as high"],
+		],
+		[
+			"nothing when maps differ only in off-level support",
+			catalogOf(
+				row({
+					id: "gpt-5.9",
+					provider: "openai",
+					api: "openai-responses",
+					reasoning: true,
+					thinkingLevelMap: { off: "none", xhigh: "xhigh" },
+				}),
+				row({
+					id: "gpt-5.9",
+					provider: "github-copilot",
+					api: "openai-responses",
+					reasoning: true,
+					thinkingLevelMap: { off: null, xhigh: "xhigh" },
+				}),
+			),
+			[],
+		],
+	])("reports %s", (_name, catalog, expected) => {
+		expect(validateModelCatalog(catalog)).toEqual(expected);
+	});
+});
diff --git a/packages/ai/test/supports-xhigh.test.ts b/packages/ai/test/supports-xhigh.test.ts
index 072043589c..2e2b763295 100644
--- a/packages/ai/test/supports-xhigh.test.ts
+++ b/packages/ai/test/supports-xhigh.test.ts
@@ -80,7 +80,7 @@ describe("getSupportedThinkingLevels", () => {
 			expect(getSupportedThinkingLevels(codexModel!)).toEqual(["off", "low", "medium", "high", "xhigh", "max"]);
 			expect(apiModel!.contextWindow).toBe(1050000);
 			expect(apiModel!.maxTokens).toBe(128000);
-			expect(codexModel!.contextWindow).toBe(272000);
+			expect(codexModel!.contextWindow).toBe(1050000);
 			expect(codexModel!.maxTokens).toBe(128000);
 		},
 	);
diff --git a/packages/coding-agent/.changes/res-1273-fixed-effort-model-default.md b/packages/coding-agent/.changes/res-1273-fixed-effort-model-default.md
new file mode 100644
index 0000000000..f33cd6cd75
--- /dev/null
+++ b/packages/coding-agent/.changes/res-1273-fixed-effort-model-default.md
@@ -0,0 +1 @@
+- Fixed switching to a model with no selectable thinking levels (e.g. Kimi K3 on Moonshot) overwriting the saved thinking default with off and leaving thinking disabled after switching away; cycling effort on such a model is now a no-op.
diff --git a/packages/coding-agent/src/core/agent-session.ts b/packages/coding-agent/src/core/agent-session.ts
index eec21e2341..bca9916b51 100644
--- a/packages/coding-agent/src/core/agent-session.ts
+++ b/packages/coding-agent/src/core/agent-session.ts
@@ -7386,7 +7386,8 @@ export class AgentSession {
 
 		if (isChanging) {
 			this.sessionManager.appendThinkingLevelChange(effectiveLevel);
-			if (this.supportsThinking() || effectiveLevel !== "off") {
+			// A forced clamp on a model with no selectable levels is not a user preference; keep the persisted default.
+			if ((this.supportsThinking() && availableLevels.length > 0) || effectiveLevel !== "off") {
 				this.settingsManager.setDefaultThinkingLevel(effectiveLevel);
 			}
 			this._emit({ type: "thinking_level_changed", level: effectiveLevel });
@@ -7445,6 +7446,7 @@ export class AgentSession {
 		if (!this.supportsThinking()) return undefined;
 
 		const levels = this.getAvailableThinkingLevels();
+		if (levels.length === 0) return undefined;
 		const currentIndex = levels.indexOf(this.thinkingLevel);
 		const nextIndex = (currentIndex + 1) % levels.length;
 		const nextLevel = levels[nextIndex];
@@ -7466,7 +7468,7 @@ export class AgentSession {
 		if (explicitLevel !== undefined) {
 			return explicitLevel;
 		}
-		if (!this.supportsThinking()) {
+		if (!this.supportsThinking() || this.getAvailableThinkingLevels().length === 0) {
 			return this.settingsManager.getDefaultThinkingLevel() ?? DEFAULT_THINKING_LEVEL;
 		}
 		return this.thinkingLevel;
diff --git a/packages/coding-agent/test/agent-session-services.test.ts b/packages/coding-agent/test/agent-session-services.test.ts
index ca25424d6d..8c6f2d74e4 100644
--- a/packages/coding-agent/test/agent-session-services.test.ts
+++ b/packages/coding-agent/test/agent-session-services.test.ts
@@ -412,4 +412,59 @@ describe("createAgentSessionFromServices", () => {
 			withMessageController.dispose();
 		}
 	});
+
+	it("keeps the persisted thinking default when a model offers no selectable levels", async () => {
+		const tempDir = join(tmpdir(), `pi-session-fixed-effort-${Date.now()}-${Math.random().toString(36).slice(2)}`);
+		mkdirSync(tempDir, { recursive: true });
+		cleanupPaths.push(tempDir);
+
+		const faux = registerFauxProvider({
+			models: [
+				{ id: "capable", reasoning: true },
+				{ id: "fixed-effort", reasoning: true },
+			],
+		});
+		unregisters.push(() => faux.unregister());
+		const [capable, fixedEffort] = faux.models;
+		// A reasoning model whose transport cannot send effort: every level nulled.
+		fixedEffort.thinkingLevelMap = { off: null, minimal: null, low: null, medium: null, high: null, xhigh: null, max: null };
+
+		const authStorage = AuthStorage.inMemory();
+		authStorage.setRuntimeApiKey(capable.provider, "faux-key");
+		const settingsManager = SettingsManager.inMemory();
+		const services = await createAgentSessionServices({
+			cwd: tempDir,
+			agentDir: tempDir,
+			authStorage,
+			settingsManager,
+			resourceLoaderOptions: { noPromptTemplates: true, noThemes: true },
+		});
+		services.modelRegistry.registerProvider(capable.provider, {
+			baseUrl: capable.baseUrl,
+			apiKey: "faux-key",
+			api: faux.api,
+			models: faux.models,
+		});
+
+		const { session } = await createAgentSessionFromServices({
+			services,
+			sessionManager: SessionManager.create(tempDir, join(tempDir, "sessions")),
+			model: capable,
+		});
+		try {
+			session.setThinkingLevel("high");
+			expect(settingsManager.getDefaultThinkingLevel()).toBe("high");
+
+			await session.setModel(fixedEffort, { waitForExtensions: false });
+			expect(session.thinkingLevel).toBe("off");
+			expect(session.cycleThinkingLevel()).toBeUndefined();
+			expect(session.thinkingLevel).toBe("off");
+			expect(settingsManager.getDefaultThinkingLevel()).toBe("high");
+
+			await session.setModel(capable, { waitForExtensions: false });
+			expect(session.thinkingLevel).toBe("high");
+		} finally {
+			session.dispose();
+		}
+	});
 });