diff --git a/open-sse/handlers/chatCore.js b/open-sse/handlers/chatCore.js
index e942bd75..7fedca56 100644
--- a/open-sse/handlers/chatCore.js
+++ b/open-sse/handlers/chatCore.js
@@ -91,7 +91,12 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
   // differ — kimi/glm only do /chat/completions). Undeclared models keep the
   // upstream default (use the transport), preserving behavior for glm/deepseek/...
   const useTransport = (!modelSupportedFormats || modelSupportedFormats.includes(sourceFormat)) ? runtimeTransport : null;
-  const targetFormat = modelTargetFormat || useTransport?.format || getTargetFormat(provider, credentials);
+  // A source-format-matched endpoint keeps the request lossless. Prefer it
+  // over a model-level targetFormat, which is only the fallback for clients
+  // whose wire format has no supported transport (for example MiniMax-M3:
+  // OpenAI clients should stay on /chat/completions; other clients can fall
+  // back to its declared Claude target).
+  const targetFormat = useTransport?.format || modelTargetFormat || getTargetFormat(provider, credentials);
   if (useTransport && credentials) credentials.runtimeTransport = useTransport;
   const stripList = getModelStrip(alias, model);
   const upstreamModel = getModelUpstreamId(alias, model);
diff --git a/tests/unit/minimax-transport-target-format.test.js b/tests/unit/minimax-transport-target-format.test.js
new file mode 100644
index 00000000..489252da
--- /dev/null
+++ b/tests/unit/minimax-transport-target-format.test.js
@@ -0,0 +1,189 @@
+/**
+ * Multi-transport providers must keep the request body and selected endpoint on
+ * the same wire format. MiniMax-M3 declares a Claude target for compatibility,
+ * but an OpenAI client should use MiniMax's matching OpenAI transport without
+ * an OpenAI -> Claude translation.
+ * Regression: https://github.com/decolua/9router/issues/3418
+ */
+import { beforeEach, describe, expect, it, vi } from "vitest";
+
+const {
+  executeMock,
+  translateRequestMock,
+  handleNonStreamingResponseMock,
+} = vi.hoisted(() => ({
+  executeMock: vi.fn(),
+  translateRequestMock: vi.fn((sourceFormat, targetFormat, model, body) => ({
+    ...body,
+    model,
+    _translatedFrom: sourceFormat,
+    _translatedTo: targetFormat,
+  })),
+  handleNonStreamingResponseMock: vi.fn(async () => ({ success: true })),
+}));
+
+vi.mock("../../open-sse/executors/index.js", () => ({
+  getExecutor: vi.fn(() => ({
+    execute: executeMock,
+    refreshCredentials: vi.fn().mockResolvedValue(null),
+  })),
+}));
+
+vi.mock("../../open-sse/translator/index.js", () => ({
+  translateRequest: translateRequestMock,
+}));
+
+vi.mock("../../open-sse/handlers/chatCore/nonStreamingHandler.js", () => ({
+  handleNonStreamingResponse: handleNonStreamingResponseMock,
+}));
+
+vi.mock("../../open-sse/utils/requestLogger.js", () => ({
+  createRequestLogger: vi.fn(async () => ({
+    logClientRawRequest: vi.fn(),
+    logRawRequest: vi.fn(),
+    logTargetRequest: vi.fn(),
+    logError: vi.fn(),
+  })),
+}));
+
+vi.mock("../../open-sse/utils/clientDetector.js", () => ({
+  detectClientTool: vi.fn(() => null),
+  isNativePassthrough: vi.fn(() => false),
+}));
+
+vi.mock("../../open-sse/utils/bypassHandler.js", () => ({
+  handleBypassRequest: vi.fn(() => null),
+}));
+
+vi.mock("../../open-sse/utils/streamHandler.js", () => ({
+  createStreamController: vi.fn(() => ({
+    signal: undefined,
+    handleComplete: vi.fn(),
+    handleError: vi.fn(),
+  })),
+}));
+
+vi.mock("../../open-sse/services/tokenRefresh.js", () => ({
+  refreshWithRetry: vi.fn(),
+}));
+
+vi.mock("../../open-sse/utils/proxyFetch.js", () => ({
+  default: vi.fn(),
+  proxyAwareFetch: vi.fn(),
+}));
+
+vi.mock("../../open-sse/translator/formats/claude.js", () => ({
+  normalizeClaudePassthrough: vi.fn(),
+  anchorClaudeCache: vi.fn(),
+}));
+
+vi.mock("../../open-sse/utils/toolDeduper.js", () => ({
+  dedupeTools: vi.fn((tools) => ({ tools, stripped: [] })),
+}));
+
+vi.mock("../../open-sse/rtk/caveman.js", () => ({ injectCaveman: vi.fn() }));
+vi.mock("../../open-sse/rtk/ponytail.js", () => ({ injectPonytail: vi.fn() }));
+vi.mock("../../open-sse/rtk/index.js", () => ({
+  compressMessages: vi.fn(() => null),
+  formatRtkLog: vi.fn(() => ""),
+}));
+vi.mock("../../open-sse/rtk/headroom.js", () => ({
+  compressWithHeadroom: vi.fn(async () => null),
+  formatHeadroomLog: vi.fn(() => ""),
+  formatHeadroomSizeLog: vi.fn(() => ""),
+  isHeadroomPhantomSavings: vi.fn(() => false),
+}));
+vi.mock("../../open-sse/rtk/pxpipe.js", () => ({
+  compressWithPxpipe: vi.fn(async () => ({ body: null, summary: null })),
+}));
+
+vi.mock("../../open-sse/translator/concerns/prefetch.js", () => ({
+  prefetchRemoteImages: vi.fn(async () => 0),
+}));
+
+vi.mock("../../open-sse/handlers/chatCore/requestDetail.js", () => ({
+  buildRequestDetail: vi.fn((detail) => detail),
+  extractRequestConfig: vi.fn((body, stream) => ({ body, stream })),
+}));
+
+vi.mock("../../open-sse/utils/error.js", () => ({
+  createErrorResult: vi.fn((status, message) => ({ success: false, status, error: message })),
+  formatProviderError: vi.fn((error) => error.message),
+  parseUpstreamError: vi.fn(),
+}));
+
+vi.mock("@/lib/usageDb.js", () => ({
+  trackPendingRequest: vi.fn(),
+  appendRequestLog: vi.fn(() => Promise.resolve()),
+  saveRequestDetail: vi.fn(() => Promise.resolve()),
+}));
+
+function makeOptions(body) {
+  return {
+    body,
+    modelInfo: { provider: "minimax-cn", model: "MiniMax-M3" },
+    credentials: { apiKey: "test-api-key", providerSpecificData: {} },
+    clientRawRequest: {
+      endpoint: "/v1/chat/completions",
+      body,
+      headers: { accept: "application/json" },
+    },
+    connectionId: "test-connection",
+    log: { debug: vi.fn(), info: vi.fn(), warn: vi.fn(), error: vi.fn() },
+  };
+}
+
+describe("MiniMax-M3 multi-transport routing", () => {
+  beforeEach(() => {
+    executeMock.mockReset();
+    translateRequestMock.mockClear();
+    handleNonStreamingResponseMock.mockClear();
+    executeMock.mockResolvedValue({
+      response: new Response("{}", {
+        status: 200,
+        headers: { "content-type": "application/json" },
+      }),
+      url: "https://api.minimaxi.com/v1/chat/completions",
+      headers: {},
+      transformedBody: {},
+    });
+  });
+
+  it("keeps OpenAI image blocks on the matching OpenAI transport", async () => {
+    const imageBlock = {
+      type: "image_url",
+      image_url: { url: "data:image/png;base64,AAAB" },
+    };
+    const body = {
+      model: "minimax-cn/MiniMax-M3",
+      stream: false,
+      messages: [{
+        role: "user",
+        content: [{ type: "text", text: "Describe this image" }, imageBlock],
+      }],
+    };
+
+    const { handleChatCore } = await import("../../open-sse/handlers/chatCore.js");
+    await handleChatCore(makeOptions(body));
+
+    expect(translateRequestMock).toHaveBeenCalledWith(
+      "openai",
+      "openai",
+      "MiniMax-M3",
+      expect.any(Object),
+      false,
+      expect.any(Object),
+      "minimax-cn",
+      expect.any(Object),
+      expect.anything(),
+      "test-connection",
+      null,
+    );
+    expect(executeMock).toHaveBeenCalledTimes(1);
+    const requestBody = executeMock.mock.calls[0][0].body;
+    expect(requestBody.messages[0].content).toContainEqual(imageBlock);
+    expect(requestBody._translatedTo).toBe("openai");
+    expect(requestBody).not.toHaveProperty("system");
+    expect(executeMock.mock.calls[0][0].credentials.runtimeTransport.format).toBe("openai");
+  });
+});