feat: add zai/glm-4.6v image understanding support (#10267)

Fixes #10265. Thanks @liuy.

CHANGELOG.md

@@ -13,6 +13,7 @@ Docs: https://docs.openclaw.ai
 - Gateway: add agent management RPC methods for the web UI (`agents.create`, `agents.update`, `agents.delete`). (#11045) Thanks @advaitpaliwal.
 - Web UI: show a Compaction divider in chat history. (#11341) Thanks @Takhoffman.
 - Agents: include runtime shell in agent envelopes. (#1835) Thanks @Takhoffman.
+- Agents: auto-select `zai/glm-4.6v` for image understanding when ZAI is primary provider. (#10267) Thanks @liuy.
 - Paths: add `OPENCLAW_HOME` for overriding the home directory used by internal path resolution. (#12091) Thanks @sebslight.
 
 ### Fixes

src/agents/tools/image-tool.test.ts

@@ -22,6 +22,8 @@ describe("image tool implicit imageModel config", () => {
     vi.stubEnv("ANTHROPIC_API_KEY", "");
     vi.stubEnv("ANTHROPIC_OAUTH_TOKEN", "");
     vi.stubEnv("MINIMAX_API_KEY", "");
+    vi.stubEnv("ZAI_API_KEY", "");
+    vi.stubEnv("Z_AI_API_KEY", "");
     // Avoid implicit Copilot provider discovery hitting the network in tests.
     vi.stubEnv("COPILOT_GITHUB_TOKEN", "");
     vi.stubEnv("GH_TOKEN", "");
@@ -58,6 +60,21 @@ describe("image tool implicit imageModel config", () => {
     expect(createImageTool({ config: cfg, agentDir })).not.toBeNull();
   });
 
+  it("pairs zai primary with glm-4.6v (and fallbacks) when auth exists", async () => {
+    const agentDir = await fs.mkdtemp(path.join(os.tmpdir(), "openclaw-image-"));
+    vi.stubEnv("ZAI_API_KEY", "zai-test");
+    vi.stubEnv("OPENAI_API_KEY", "openai-test");
+    vi.stubEnv("ANTHROPIC_API_KEY", "anthropic-test");
+    const cfg: OpenClawConfig = {
+      agents: { defaults: { model: { primary: "zai/glm-4.7" } } },
+    };
+    expect(resolveImageModelConfigForTool({ cfg, agentDir })).toEqual({
+      primary: "zai/glm-4.6v",
+      fallbacks: ["openai/gpt-5-mini", "anthropic/claude-opus-4-5"],
+    });
+    expect(createImageTool({ config: cfg, agentDir })).not.toBeNull();
+  });
+
   it("pairs a custom provider when it declares an image-capable model", async () => {
     const agentDir = await fs.mkdtemp(path.join(os.tmpdir(), "openclaw-image-"));
     await writeAuthProfiles(agentDir, {

src/agents/tools/image-tool.ts

@@ -116,6 +116,8 @@ export function resolveImageModelConfigForTool(params: {
     preferred = "minimax/MiniMax-VL-01";
   } else if (providerOk && providerVisionFromConfig) {
     preferred = providerVisionFromConfig;
+  } else if (primary.provider === "zai" && providerOk) {
+    preferred = "zai/glm-4.6v";
   } else if (primary.provider === "openai" && openaiOk) {
     preferred = "openai/gpt-5-mini";
   } else if (primary.provider === "anthropic" && anthropicOk) {

src/media-understanding/defaults.ts

@@ -32,5 +32,22 @@ export const DEFAULT_AUDIO_MODELS: Record<string, string> = {
   openai: "gpt-4o-mini-transcribe",
   deepgram: "nova-3",
 };
+
+export const AUTO_AUDIO_KEY_PROVIDERS = ["openai", "groq", "deepgram", "google"] as const;
+export const AUTO_IMAGE_KEY_PROVIDERS = [
+  "openai",
+  "anthropic",
+  "google",
+  "minimax",
+  "zai",
+] as const;
+export const AUTO_VIDEO_KEY_PROVIDERS = ["google"] as const;
+export const DEFAULT_IMAGE_MODELS: Record<string, string> = {
+  openai: "gpt-5-mini",
+  anthropic: "claude-opus-4-6",
+  google: "gemini-3-flash-preview",
+  minimax: "MiniMax-VL-01",
+  zai: "glm-4.6v",
+};
 export const CLI_OUTPUT_MAX_BUFFER = 5 * MB;
 export const DEFAULT_MEDIA_CONCURRENCY = 2;

src/media-understanding/providers/index.ts

@@ -6,6 +6,7 @@ import { googleProvider } from "./google/index.js";
 import { groqProvider } from "./groq/index.js";
 import { minimaxProvider } from "./minimax/index.js";
 import { openaiProvider } from "./openai/index.js";
+import { zaiProvider } from "./zai/index.js";
 
 const PROVIDERS: MediaUnderstandingProvider[] = [
   groqProvider,
@@ -13,6 +14,7 @@ const PROVIDERS: MediaUnderstandingProvider[] = [
   googleProvider,
   anthropicProvider,
   minimaxProvider,
+  zaiProvider,
   deepgramProvider,
 ];
 

src/media-understanding/providers/zai/index.ts

@@ -0,0 +1,8 @@
+import type { MediaUnderstandingProvider } from "../../types.js";
+import { describeImageWithModel } from "../image.js";
+
+export const zaiProvider: MediaUnderstandingProvider = {
+  id: "zai",
+  capabilities: ["image"],
+  describeImage: describeImageWithModel,
+};

src/media-understanding/runner.ts

@@ -27,8 +27,12 @@ import { logVerbose, shouldLogVerbose } from "../globals.js";
 import { runExec } from "../process/exec.js";
 import { MediaAttachmentCache, normalizeAttachments, selectAttachments } from "./attachments.js";
 import {
+  AUTO_AUDIO_KEY_PROVIDERS,
+  AUTO_IMAGE_KEY_PROVIDERS,
+  AUTO_VIDEO_KEY_PROVIDERS,
   CLI_OUTPUT_MAX_BUFFER,
   DEFAULT_AUDIO_MODELS,
+  DEFAULT_IMAGE_MODELS,
   DEFAULT_TIMEOUT_SECONDS,
 } from "./defaults.js";
 import { isMediaUnderstandingSkipError, MediaUnderstandingSkipError } from "./errors.js";
@@ -48,16 +52,6 @@ import {
 } from "./resolve.js";
 import { estimateBase64Size, resolveVideoMaxBase64Bytes } from "./video.js";
 
-const AUTO_AUDIO_KEY_PROVIDERS = ["openai", "groq", "deepgram", "google"] as const;
-const AUTO_IMAGE_KEY_PROVIDERS = ["openai", "anthropic", "google", "minimax"] as const;
-const AUTO_VIDEO_KEY_PROVIDERS = ["google"] as const;
-const DEFAULT_IMAGE_MODELS: Record<string, string> = {
-  openai: "gpt-5-mini",
-  anthropic: "claude-opus-4-6",
-  google: "gemini-3-flash-preview",
-  minimax: "MiniMax-VL-01",
-};
-
 export type ActiveMediaModel = {
   provider: string;
   model?: string;