diff --git a/packages/ai/CHANGELOG.md b/packages/ai/CHANGELOG.md
index 948bcb32..6966d94a 100644
--- a/packages/ai/CHANGELOG.md
+++ b/packages/ai/CHANGELOG.md
@@ -11,6 +11,8 @@
 
 - Fixed Anthropic-compatible replay for providers that return empty thinking signatures by adding an opt-in `allowEmptySignature` compatibility flag ([#4464](https://github.com/earendil-works/pi/issues/4464)).
 - Fixed OpenAI and OpenRouter GPT-5.5 Pro thinking level metadata to expose only supported medium, high, and xhigh efforts.
+- Fixed OpenCode Go Kimi K2.6 thinking-off requests to send `thinking: "none"` ([#5078](https://github.com/earendil-works/pi/issues/5078)).
+- Fixed Xiaomi Token Plan model metadata to omit unsupported `mimo-v2-flash` variants ([#5075](https://github.com/earendil-works/pi/issues/5075)).
 
 ## [0.76.0] - 2026-05-27
 
diff --git a/packages/ai/scripts/generate-models.ts b/packages/ai/scripts/generate-models.ts
index 927401f7..fc4e2c3f 100644
--- a/packages/ai/scripts/generate-models.ts
+++ b/packages/ai/scripts/generate-models.ts
@@ -269,6 +269,9 @@ function applyThinkingLevelMetadata(model: Model<any>): void {
 		// Pi's low/medium/high pass through verbatim; OpenRouter normalizes to Mercury's vocabulary.
 		mergeThinkingLevelMap(model, { off: null });
 	}
+	if (model.provider === "opencode-go" && model.id === "kimi-k2.6") {
+		mergeThinkingLevelMap(model, { off: "none" });
+	}
 }
 
 function getAnthropicMessagesCompat(provider: string, modelId: string): AnthropicMessagesCompat | undefined {
@@ -900,6 +903,9 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
 						api = "openai-completions";
 						baseUrl = `${variant.basePath}/v1`;
 					}
+					if (modelId === "kimi-k2.6") {
+						compat = { ...(compat ?? {}), thinkingFormat: "string-thinking" };
+					}
 					if (modelId === "qwen3.5-plus" || modelId === "qwen3.6-plus") {
 						api = "openai-completions";
 						baseUrl = `${variant.basePath}/v1`;
@@ -1116,6 +1122,7 @@ async function loadModelsDevData(): Promise<Model<any>[]> {
 				for (const [modelId, model] of Object.entries(data.xiaomi.models)) {
 					const m = model as ModelsDevModel;
 					if (m.tool_call !== true) continue;
+					if (provider.startsWith("xiaomi-token-plan-") && modelId === "mimo-v2-flash") continue;
 
 					models.push({
 						id: modelId,
diff --git a/packages/ai/src/models.generated.ts b/packages/ai/src/models.generated.ts
index 1996cb02..8d95054a 100644
--- a/packages/ai/src/models.generated.ts
+++ b/packages/ai/src/models.generated.ts
@@ -7968,7 +7968,9 @@ export const MODELS = {
 			api: "openai-completions",
 			provider: "opencode-go",
 			baseUrl: "https://opencode.ai/zen/go/v1",
+			compat: {"thinkingFormat":"string-thinking"},
 			reasoning: true,
+			thinkingLevelMap: {"off":"none"},
 			input: ["text", "image"],
 			cost: {
 				input: 0.95,
@@ -15929,24 +15931,6 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 	},
 	"xiaomi-token-plan-ams": {
-		"mimo-v2-flash": {
-			id: "mimo-v2-flash",
-			name: "MiMo-V2-Flash",
-			api: "openai-completions",
-			provider: "xiaomi-token-plan-ams",
-			baseUrl: "https://token-plan-ams.xiaomimimo.com/v1",
-			compat: {"requiresReasoningContentOnAssistantMessages":true,"thinkingFormat":"deepseek"},
-			reasoning: true,
-			input: ["text"],
-			cost: {
-				input: 0.1,
-				output: 0.3,
-				cacheRead: 0.01,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 65536,
-		} satisfies Model<"openai-completions">,
 		"mimo-v2-omni": {
 			id: "mimo-v2-omni",
 			name: "MiMo-V2-Omni",
@@ -16021,24 +16005,6 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 	},
 	"xiaomi-token-plan-cn": {
-		"mimo-v2-flash": {
-			id: "mimo-v2-flash",
-			name: "MiMo-V2-Flash",
-			api: "openai-completions",
-			provider: "xiaomi-token-plan-cn",
-			baseUrl: "https://token-plan-cn.xiaomimimo.com/v1",
-			compat: {"requiresReasoningContentOnAssistantMessages":true,"thinkingFormat":"deepseek"},
-			reasoning: true,
-			input: ["text"],
-			cost: {
-				input: 0.1,
-				output: 0.3,
-				cacheRead: 0.01,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 65536,
-		} satisfies Model<"openai-completions">,
 		"mimo-v2-omni": {
 			id: "mimo-v2-omni",
 			name: "MiMo-V2-Omni",
@@ -16113,24 +16079,6 @@ export const MODELS = {
 		} satisfies Model<"openai-completions">,
 	},
 	"xiaomi-token-plan-sgp": {
-		"mimo-v2-flash": {
-			id: "mimo-v2-flash",
-			name: "MiMo-V2-Flash",
-			api: "openai-completions",
-			provider: "xiaomi-token-plan-sgp",
-			baseUrl: "https://token-plan-sgp.xiaomimimo.com/v1",
-			compat: {"requiresReasoningContentOnAssistantMessages":true,"thinkingFormat":"deepseek"},
-			reasoning: true,
-			input: ["text"],
-			cost: {
-				input: 0.1,
-				output: 0.3,
-				cacheRead: 0.01,
-				cacheWrite: 0,
-			},
-			contextWindow: 262144,
-			maxTokens: 65536,
-		} satisfies Model<"openai-completions">,
 		"mimo-v2-omni": {
 			id: "mimo-v2-omni",
 			name: "MiMo-V2-Omni",
diff --git a/packages/ai/src/providers/openai-completions.ts b/packages/ai/src/providers/openai-completions.ts
index 0d37b955..2e243fb7 100644
--- a/packages/ai/src/providers/openai-completions.ts
+++ b/packages/ai/src/providers/openai-completions.ts
@@ -594,6 +594,13 @@ function buildParams(
 		if (options?.reasoningEffort && compat.supportsReasoningEffort) {
 			togetherParams.reasoning_effort = model.thinkingLevelMap?.[options.reasoningEffort] ?? options.reasoningEffort;
 		}
+	} else if (compat.thinkingFormat === "string-thinking" && model.reasoning) {
+		const stringThinkingParams = params as typeof params & { thinking?: string };
+		if (options?.reasoningEffort) {
+			stringThinkingParams.thinking = model.thinkingLevelMap?.[options.reasoningEffort] ?? options.reasoningEffort;
+		} else if (model.thinkingLevelMap?.off !== null) {
+			stringThinkingParams.thinking = model.thinkingLevelMap?.off ?? "none";
+		}
 	} else if (options?.reasoningEffort && model.reasoning && compat.supportsReasoningEffort) {
 		// OpenAI-style reasoning_effort
 		(params as any).reasoning_effort = model.thinkingLevelMap?.[options.reasoningEffort] ?? options.reasoningEffort;
diff --git a/packages/ai/src/types.ts b/packages/ai/src/types.ts
index 4345e3bf..bf997317 100644
--- a/packages/ai/src/types.ts
+++ b/packages/ai/src/types.ts
@@ -387,8 +387,16 @@ export interface OpenAICompletionsCompat {
 	requiresThinkingAsText?: boolean;
 	/** Whether all replayed assistant messages must include an empty reasoning_content field when reasoning is enabled. Default: auto-detected from URL. */
 	requiresReasoningContentOnAssistantMessages?: boolean;
-	/** Format for reasoning/thinking parameter. "openai" uses reasoning_effort, "openrouter" uses reasoning: { effort }, "deepseek" uses thinking: { type } plus reasoning_effort, "together" uses reasoning: { enabled } plus reasoning_effort when supported, "zai" uses top-level enable_thinking: boolean, "qwen" uses top-level enable_thinking: boolean, and "qwen-chat-template" uses chat_template_kwargs.enable_thinking. Default: "openai". */
-	thinkingFormat?: "openai" | "openrouter" | "deepseek" | "together" | "zai" | "qwen" | "qwen-chat-template";
+	/** Format for reasoning/thinking parameter. "openai" uses reasoning_effort, "openrouter" uses reasoning: { effort }, "deepseek" uses thinking: { type } plus reasoning_effort, "together" uses reasoning: { enabled } plus reasoning_effort when supported, "zai" uses top-level enable_thinking: boolean, "qwen" uses top-level enable_thinking: boolean, "qwen-chat-template" uses chat_template_kwargs.enable_thinking, and "string-thinking" uses top-level thinking: string. Default: "openai". */
+	thinkingFormat?:
+		| "openai"
+		| "openrouter"
+		| "deepseek"
+		| "together"
+		| "zai"
+		| "qwen"
+		| "qwen-chat-template"
+		| "string-thinking";
 	/** OpenRouter-specific routing preferences. Only used when baseUrl points to OpenRouter. */
 	openRouterRouting?: OpenRouterRouting;
 	/** Vercel AI Gateway routing preferences. Only used when baseUrl points to Vercel AI Gateway. */
diff --git a/packages/ai/test/openai-completions-tool-choice.test.ts b/packages/ai/test/openai-completions-tool-choice.test.ts
index ec98427b..66cfb562 100644
--- a/packages/ai/test/openai-completions-tool-choice.test.ts
+++ b/packages/ai/test/openai-completions-tool-choice.test.ts
@@ -995,6 +995,51 @@ describe("openai-completions tool_choice", () => {
 		expect(messages[0]).not.toHaveProperty("reasoning");
 	});
 
+	it("sends thinking none for OpenCode Go Kimi K2.6 when thinking is off", async () => {
+		const model = getModel("opencode-go", "kimi-k2.6")!;
+		let payload: unknown;
+
+		await streamSimple(
+			model,
+			{
+				messages: [{ role: "user", content: "Hi", timestamp: Date.now() }],
+			},
+			{
+				apiKey: "test",
+				onPayload: (params: unknown) => {
+					payload = params;
+				},
+			},
+		).result();
+
+		const params = (payload ?? mockState.lastParams) as { thinking?: string; reasoning_effort?: string };
+		expect(params.thinking).toBe("none");
+		expect(params.reasoning_effort).toBeUndefined();
+	});
+
+	it("sends thinking effort for OpenCode Go Kimi K2.6 when thinking is enabled", async () => {
+		const model = getModel("opencode-go", "kimi-k2.6")!;
+		let payload: unknown;
+
+		await streamSimple(
+			model,
+			{
+				messages: [{ role: "user", content: "Hi", timestamp: Date.now() }],
+			},
+			{
+				apiKey: "test",
+				reasoning: "high",
+				onPayload: (params: unknown) => {
+					payload = params;
+				},
+			},
+		).result();
+
+		const params = (payload ?? mockState.lastParams) as { thinking?: string; reasoning_effort?: string };
+		expect(params.thinking).toBe("high");
+		expect(params.reasoning_effort).toBeUndefined();
+	});
+
 	it("does not double-count reasoning tokens in completion usage", async () => {
 		mockState.chunks = [
 			{
diff --git a/packages/ai/test/xiaomi-models.test.ts b/packages/ai/test/xiaomi-models.test.ts
new file mode 100644
index 00000000..6fb1f4b8
--- /dev/null
+++ b/packages/ai/test/xiaomi-models.test.ts
@@ -0,0 +1,15 @@
+import { describe, expect, it } from "vitest";
+import { getModel, getModels } from "../src/models.ts";
+
+describe("Xiaomi MiMo models", () => {
+	it("keeps mimo-v2-flash on the API billing provider", () => {
+		expect(getModel("xiaomi", "mimo-v2-flash")).toBeDefined();
+	});
+
+	it.each(["xiaomi-token-plan-cn", "xiaomi-token-plan-ams", "xiaomi-token-plan-sgp"] as const)(
+		"omits mimo-v2-flash from %s",
+		(provider) => {
+			expect(getModels(provider).some((model) => model.id === "mimo-v2-flash")).toBe(false);
+		},
+	);
+});