From 535fdcea2c6a47251eb677c652ea3457d98a8190 Mon Sep 17 00:00:00 2001
From: Mario Zechner <badlogicgames@gmail.com>
Date: Thu, 23 Apr 2026 21:20:23 +0200
Subject: [PATCH] test(ai): refresh responses and codex e2e coverage closes
 #3576

---
 packages/ai/CHANGELOG.md        |  1 +
 packages/ai/test/stream.test.ts | 19 +++++++++++--------
 2 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/packages/ai/CHANGELOG.md b/packages/ai/CHANGELOG.md
index add6f587..d118b3d5 100644
--- a/packages/ai/CHANGELOG.md
+++ b/packages/ai/CHANGELOG.md
@@ -9,6 +9,7 @@
 ### Fixed
 
 - Fixed `openai-completions` streamed tool-call assembly to coalesce deltas by stable tool index when OpenAI-compatible gateways mutate tool call IDs mid-stream, preventing malformed Kimi K2.6/OpenCode tool streams from splitting one call into multiple bogus tool calls ([#3576](https://github.com/badlogic/pi-mono/issues/3576))
+- Fixed `packages/ai` E2E coverage to use currently supported OpenAI Responses and OpenAI Codex models, and updated the Bedrock adaptive-thinking payload expectation to match the current `display: "summarized"` shape.
 - Fixed built-in `kimi-coding` model generation to attach `User-Agent: KimiCLI/1.5` to all generated Kimi models, overriding the Anthropic SDK default UA so direct Kimi Coding requests use the provider's expected client identity ([#3586](https://github.com/badlogic/pi-mono/issues/3586))
 
 ## [0.69.0] - 2026-04-22
diff --git a/packages/ai/test/stream.test.ts b/packages/ai/test/stream.test.ts
index 4bfdedc7..432a47c7 100644
--- a/packages/ai/test/stream.test.ts
+++ b/packages/ai/test/stream.test.ts
@@ -447,8 +447,8 @@ describe("Generate E2E Tests", () => {
 		});
 	});
 
-	describe.skipIf(!process.env.OPENAI_API_KEY)("OpenAI Responses Provider (gpt-5-mini)", () => {
-		const llm = getModel("openai", "gpt-5-mini");
+	describe.skipIf(!process.env.OPENAI_API_KEY)("OpenAI Responses Provider (gpt-5.4)", () => {
+		const llm = getModel("openai", "gpt-5.4");
 
 		it("should complete basic text generation", { retry: 3 }, async () => {
 			await basicTextGeneration(llm);
@@ -1120,8 +1120,8 @@ describe("Generate E2E Tests", () => {
 		});
 	});
 
-	describe("OpenAI Codex Provider (gpt-5.3-codex)", () => {
-		const llm = getModel("openai-codex", "gpt-5.3-codex");
+	describe("OpenAI Codex Provider (gpt-5.5)", () => {
+		const llm = getModel("openai-codex", "gpt-5.5");
 
 		it.skipIf(!openaiCodexToken)("should complete basic text generation", { retry: 3 }, async () => {
 			await basicTextGeneration(llm, { apiKey: openaiCodexToken });
@@ -1148,8 +1148,8 @@ describe("Generate E2E Tests", () => {
 		});
 	});
 
-	describe("OpenAI Codex Provider (gpt-5.3-codex via WebSocket)", () => {
-		const llm = getModel("openai-codex", "gpt-5.3-codex");
+	describe("OpenAI Codex Provider (gpt-5.5 via WebSocket)", () => {
+		const llm = getModel("openai-codex", "gpt-5.5");
 		const wsOptions = { apiKey: openaiCodexToken, transport: "websocket" as const };
 
 		it.skipIf(!openaiCodexToken)("should complete basic text generation", { retry: 3 }, async () => {
@@ -1237,13 +1237,16 @@ describe("Generate E2E Tests", () => {
 
 			const payload = capturedPayload as {
 				additionalModelRequestFields?: {
-					thinking?: { type?: string };
+					thinking?: { type?: string; display?: string };
 					output_config?: { effort?: string };
 					anthropic_beta?: string[];
 				};
 			};
 
-			expect(payload.additionalModelRequestFields?.thinking).toEqual({ type: "adaptive" });
+			expect(payload.additionalModelRequestFields?.thinking).toEqual({
+				type: "adaptive",
+				display: "summarized",
+			});
 			expect(payload.additionalModelRequestFields?.output_config).toEqual({ effort: "max" });
 			expect(payload.additionalModelRequestFields?.anthropic_beta).toBeUndefined();
 		});