abort route

packages/ai/src/models.generated.ts

@@ -9395,7 +9395,7 @@ export const MODELS = {
         cacheWrite: 0,
       },
       contextWindow: 262144,
-      maxTokens: 65536,
+      maxTokens: 32000,
     } satisfies Model<"openai-completions">,
     "openrouter/hunter-alpha": {
       id: "openrouter/hunter-alpha",
@@ -9412,7 +9412,7 @@ export const MODELS = {
         cacheWrite: 0,
       },
       contextWindow: 1048576,
-      maxTokens: 65536,
+      maxTokens: 32000,
     } satisfies Model<"openai-completions">,
     "prime-intellect/intellect-3": {
       id: "prime-intellect/intellect-3",
@@ -10451,6 +10451,23 @@ export const MODELS = {
       contextWindow: 2000000,
       maxTokens: 30000,
     } satisfies Model<"openai-completions">,
+    "x-ai/grok-4.20-beta": {
+      id: "x-ai/grok-4.20-beta",
+      name: "xAI: Grok 4.20 Beta",
+      api: "openai-completions",
+      provider: "openrouter",
+      baseUrl: "https://openrouter.ai/api/v1",
+      reasoning: true,
+      input: ["text", "image"],
+      cost: {
+        input: 2,
+        output: 6,
+        cacheRead: 0.19999999999999998,
+        cacheWrite: 0,
+      },
+      contextWindow: 2000000,
+      maxTokens: 4096,
+    } satisfies Model<"openai-completions">,
     "x-ai/grok-code-fast-1": {
       id: "x-ai/grok-code-fast-1",
       name: "xAI: Grok Code Fast 1",
@@ -12793,7 +12810,7 @@ export const MODELS = {
       cost: {
         input: 3,
         output: 15,
-        cacheRead: 0,
+        cacheRead: 0.75,
         cacheWrite: 0,
       },
       contextWindow: 256000,
@@ -12867,6 +12884,57 @@ export const MODELS = {
       contextWindow: 2000000,
       maxTokens: 30000,
     } satisfies Model<"anthropic-messages">,
+    "xai/grok-4.20-multi-agent-beta": {
+      id: "xai/grok-4.20-multi-agent-beta",
+      name: "Grok 4.20 Multi Agent Beta",
+      api: "anthropic-messages",
+      provider: "vercel-ai-gateway",
+      baseUrl: "https://ai-gateway.vercel.sh",
+      reasoning: true,
+      input: ["text"],
+      cost: {
+        input: 2,
+        output: 6,
+        cacheRead: 0.19999999999999998,
+        cacheWrite: 0,
+      },
+      contextWindow: 2000000,
+      maxTokens: 2000000,
+    } satisfies Model<"anthropic-messages">,
+    "xai/grok-4.20-non-reasoning-beta": {
+      id: "xai/grok-4.20-non-reasoning-beta",
+      name: "Grok 4.20 Beta Non-Reasoning",
+      api: "anthropic-messages",
+      provider: "vercel-ai-gateway",
+      baseUrl: "https://ai-gateway.vercel.sh",
+      reasoning: false,
+      input: ["text", "image"],
+      cost: {
+        input: 2,
+        output: 6,
+        cacheRead: 0.19999999999999998,
+        cacheWrite: 0,
+      },
+      contextWindow: 2000000,
+      maxTokens: 2000000,
+    } satisfies Model<"anthropic-messages">,
+    "xai/grok-4.20-reasoning-beta": {
+      id: "xai/grok-4.20-reasoning-beta",
+      name: "Grok 4.20 Beta Reasoning",
+      api: "anthropic-messages",
+      provider: "vercel-ai-gateway",
+      baseUrl: "https://ai-gateway.vercel.sh",
+      reasoning: true,
+      input: ["text", "image"],
+      cost: {
+        input: 2,
+        output: 6,
+        cacheRead: 0.19999999999999998,
+        cacheWrite: 0,
+      },
+      contextWindow: 2000000,
+      maxTokens: 2000000,
+    } satisfies Model<"anthropic-messages">,
     "xai/grok-code-fast-1": {
       id: "xai/grok-code-fast-1",
       name: "Grok Code Fast 1",
@@ -13396,9 +13464,9 @@ export const MODELS = {
       contextWindow: 2000000,
       maxTokens: 30000,
     } satisfies Model<"openai-completions">,
-    "grok-4.20-experimental-beta-0304-non-reasoning": {
+    "grok-4.20-beta-latest-non-reasoning": {
-      id: "grok-4.20-experimental-beta-0304-non-reasoning",
+      id: "grok-4.20-beta-latest-non-reasoning",
-      name: "Grok 4.20 (Experimental, Non-Reasoning)",
+      name: "Grok 4.20 Beta (Non-Reasoning)",
       api: "openai-completions",
       provider: "xai",
       baseUrl: "https://api.x.ai/v1",
@@ -13413,9 +13481,9 @@ export const MODELS = {
       contextWindow: 2000000,
       maxTokens: 30000,
     } satisfies Model<"openai-completions">,
-    "grok-4.20-experimental-beta-0304-reasoning": {
+    "grok-4.20-beta-latest-reasoning": {
-      id: "grok-4.20-experimental-beta-0304-reasoning",
+      id: "grok-4.20-beta-latest-reasoning",
-      name: "Grok 4.20 (Experimental, Reasoning)",
+      name: "Grok 4.20 Beta (Reasoning)",
       api: "openai-completions",
       provider: "xai",
       baseUrl: "https://api.x.ai/v1",
@@ -13430,9 +13498,9 @@ export const MODELS = {
       contextWindow: 2000000,
       maxTokens: 30000,
     } satisfies Model<"openai-completions">,
-    "grok-4.20-multi-agent-experimental-beta-0304": {
+    "grok-4.20-multi-agent-beta-latest": {
-      id: "grok-4.20-multi-agent-experimental-beta-0304",
+      id: "grok-4.20-multi-agent-beta-latest",
-      name: "Grok 4.20 Multi-Agent (Experimental)",
+      name: "Grok 4.20 Multi-Agent Beta",
       api: "openai-completions",
       provider: "xai",
       baseUrl: "https://api.x.ai/v1",

packages/coding-agent/src/core/gateway/runtime.ts

@@ -282,9 +282,20 @@ export class GatewayRuntime {
 
   abortSession(sessionKey: string): boolean {
     const managedSession = this.sessions.get(sessionKey);
-    if (!managedSession?.processing) {
+    if (!managedSession) {
       return false;
     }
+
+    const hadQueuedMessages = managedSession.queue.length > 0;
+    if (hadQueuedMessages) {
+      this.rejectQueuedMessages(managedSession, "Session aborted");
+      this.emitState(managedSession);
+    }
+
+    if (!managedSession.processing) {
+      return hadQueuedMessages;
+    }
+
     void managedSession.session.abort().catch((error) => {
       this.emit(managedSession, {
         type: "error",

packages/coding-agent/test/gateway-steer.test.ts

@@ -1,4 +1,5 @@
 import { describe, expect, it, vi } from "vitest";
+import type { ManagedGatewaySession } from "../src/core/gateway/internal-types.js";
 import { GatewayRuntime } from "../src/core/gateway/runtime.js";
 
 function createMockSession() {
@@ -43,9 +44,9 @@ function addManagedSession(
   session: ReturnType<typeof createMockSession>,
   processing: boolean,
 ) {
-  const managedSession = {
+  const managedSession: ManagedGatewaySession = {
     sessionKey,
-    session,
+    session: session as never,
     queue: [],
     processing,
     activeAssistantMessage: null,
@@ -124,4 +125,36 @@ describe("GatewayRuntime steer handling", () => {
       });
     });
   });
+
+  it("abort clears queued follow-ups before aborting the active session", () => {
+    const session = createMockSession();
+    const runtime = createRuntime(session);
+    const managedSession = addManagedSession(runtime, "chat", session, true);
+    const resolve = vi.fn();
+
+    managedSession.queue.push({
+      request: {
+        sessionKey: "chat",
+        text: "stale follow-up",
+        source: "extension",
+      },
+      resolve,
+    });
+
+    const result = (
+      runtime as unknown as {
+        abortSession: (sessionKey: string) => boolean;
+      }
+    ).abortSession("chat");
+
+    expect(result).toBe(true);
+    expect(managedSession.queue).toHaveLength(0);
+    expect(resolve).toHaveBeenCalledWith({
+      ok: false,
+      response: "",
+      error: "Session aborted",
+      sessionKey: "chat",
+    });
+    expect(session.abort).toHaveBeenCalledTimes(1);
+  });
 });