Files
openclaw/extensions/qwen/provider-catalog.test.ts
Oliver Mee ae11ea5ba9 feat(qwen): add Token Plan (Team Edition) provider (#94419)
* feat(qwen): add Token Plan provider

Co-authored-by: Oliver Mee <102673257+Omee11@users.noreply.github.com>

* docs: regenerate documentation map

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
Co-authored-by: Oliver Mee <102673257+Omee11@users.noreply.github.com>
2026-07-09 22:07:01 +01:00

187 lines
7.2 KiB
TypeScript

// Qwen tests cover provider catalog plugin behavior.
import { describe, expect, it } from "vitest";
import {
applyQwenNativeStreamingUsageCompat,
buildQwenProvider,
buildQwenTokenPlanProvider,
QWEN_BASE_URL,
QWEN_36_FLASH_MODEL_ID,
QWEN_37_MAX_MODEL_ID,
QWEN_37_PLUS_MODEL_ID,
QWEN_STANDARD_GLOBAL_BASE_URL,
QWEN_DEFAULT_MODEL_ID,
QWEN_TOKEN_PLAN_CN_BASE_URL,
QWEN_TOKEN_PLAN_DEFAULT_MODEL_ID,
QWEN_TOKEN_PLAN_GLOBAL_BASE_URL,
resolveQwenTokenPlanBaseUrl,
} from "./api.js";
import manifest from "./openclaw.plugin.json" with { type: "json" };
import { buildQwenOAuthProvider } from "./provider-catalog.js";
type QwenProvider = ReturnType<typeof buildQwenProvider>;
function getQwenModelIds(provider: QwenProvider): string[] {
return provider.models.map((model) => model.id);
}
describe("qwen provider catalog", () => {
it("builds the bundled Qwen provider defaults", () => {
const provider = buildQwenProvider();
expect(provider.baseUrl).toBe(QWEN_BASE_URL);
expect(provider.api).toBe("openai-completions");
const modelIds = getQwenModelIds(provider);
expect(modelIds.length).toBeGreaterThan(0);
expect(modelIds).toContain(QWEN_DEFAULT_MODEL_ID);
expect(modelIds).not.toContain(QWEN_36_FLASH_MODEL_ID);
expect(modelIds).toContain("qwen3.6-plus");
expect(modelIds).not.toContain(QWEN_37_MAX_MODEL_ID);
expect(modelIds).toContain(QWEN_37_PLUS_MODEL_ID);
});
it("only advertises Standard-only Qwen models on Standard endpoints", () => {
const coding = buildQwenProvider({ baseUrl: QWEN_BASE_URL });
const codingTrailingDot = buildQwenProvider({
baseUrl: " https://coding-intl.dashscope.aliyuncs.com./v1 ",
});
const standard = buildQwenProvider({ baseUrl: QWEN_STANDARD_GLOBAL_BASE_URL });
expect(getQwenModelIds(coding)).toContain("qwen3.6-plus");
expect(getQwenModelIds(codingTrailingDot)).toContain("qwen3.6-plus");
expect(getQwenModelIds(standard)).toContain("qwen3.6-plus");
expect(getQwenModelIds(coding)).not.toContain(QWEN_36_FLASH_MODEL_ID);
expect(getQwenModelIds(codingTrailingDot)).not.toContain(QWEN_36_FLASH_MODEL_ID);
expect(getQwenModelIds(coding)).not.toContain(QWEN_37_MAX_MODEL_ID);
expect(getQwenModelIds(coding)).toContain(QWEN_37_PLUS_MODEL_ID);
expect(coding.models.find((model) => model.id === "qwen3.6-plus")?.reasoning).toBe(true);
expect(standard.models.find((model) => model.id === QWEN_36_FLASH_MODEL_ID)).toMatchObject({
reasoning: true,
input: ["text", "image"],
contextWindow: 1_000_000,
maxTokens: 65_536,
});
expect(standard.models.find((model) => model.id === QWEN_37_MAX_MODEL_ID)).toMatchObject({
reasoning: true,
input: ["text"],
contextWindow: 1_000_000,
maxTokens: 65_536,
});
expect(standard.models.find((model) => model.id === QWEN_37_PLUS_MODEL_ID)).toMatchObject({
reasoning: true,
input: ["text", "image"],
contextWindow: 1_000_000,
maxTokens: 65_536,
});
});
it("keeps unsupported Qwen models out of the portal catalog", () => {
const portal = buildQwenOAuthProvider();
const portalQwen36 = portal.models.find((model) => model.id === "qwen3.6-plus");
const manifestQwen36 = manifest.modelCatalog.providers["qwen-oauth"].models.find(
(model) => model.id === "qwen3.6-plus",
);
expect(getQwenModelIds(portal)).not.toContain(QWEN_36_FLASH_MODEL_ID);
expect(
manifest.modelCatalog.providers["qwen-oauth"].models.map((model) => model.id),
).not.toContain(QWEN_36_FLASH_MODEL_ID);
expect(getQwenModelIds(portal)).not.toContain(QWEN_37_MAX_MODEL_ID);
expect(getQwenModelIds(portal)).not.toContain(QWEN_37_PLUS_MODEL_ID);
expect(portalQwen36?.reasoning).toBe(true);
expect(manifestQwen36?.reasoning).toBe(portalQwen36?.reasoning);
});
it("opts native Qwen baseUrls into streaming usage only inside the extension", () => {
const nativeProvider = applyQwenNativeStreamingUsageCompat(buildQwenProvider());
expect(nativeProvider.models.length).toBeGreaterThan(0);
expect(
nativeProvider.models.every((model) => {
if (!model.compat) {
throw new Error(`expected Qwen model ${model.id} compat`);
}
return model.compat.supportsUsageInStreaming === true;
}),
).toBe(true);
const customProvider = applyQwenNativeStreamingUsageCompat({
...buildQwenProvider(),
baseUrl: "https://proxy.example.com/v1",
});
expect(
customProvider.models.some(
(model) => model.compat && model.compat.supportsUsageInStreaming === true,
),
).toBe(false);
});
});
describe("qwen token plan provider catalog", () => {
it("ships the exact 14-model Global catalog through manifest and runtime", () => {
const provider = buildQwenTokenPlanProvider();
const modelIds = provider.models.map((model) => model.id);
expect(provider.baseUrl).toBe(QWEN_TOKEN_PLAN_GLOBAL_BASE_URL);
expect(provider.api).toBe("openai-completions");
expect(modelIds).toEqual([
"qwen3.7-max",
QWEN_TOKEN_PLAN_DEFAULT_MODEL_ID,
"qwen3.6-plus",
"qwen3.6-flash",
"deepseek-v4-pro",
"deepseek-v4-flash",
"deepseek-v3.2",
"kimi-k2.7-code",
"kimi-k2.6",
"kimi-k2.5",
"glm-5.2",
"glm-5.1",
"glm-5",
"MiniMax-M2.5",
]);
expect(provider.models.every((model) => model.reasoning)).toBe(true);
expect(manifest.modelCatalog.providers["qwen-token-plan"].models).toEqual(provider.models);
expect(manifest.modelCatalog.discovery["qwen-token-plan"]).toBe("static");
});
it("uses region-scoped endpoints with the documented GLM 5.2 window", () => {
expect(resolveQwenTokenPlanBaseUrl("global")).toBe(QWEN_TOKEN_PLAN_GLOBAL_BASE_URL);
expect(resolveQwenTokenPlanBaseUrl("cn")).toBe(QWEN_TOKEN_PLAN_CN_BASE_URL);
const globalProvider = buildQwenTokenPlanProvider();
const cnProvider = buildQwenTokenPlanProvider({ baseUrl: QWEN_TOKEN_PLAN_CN_BASE_URL });
expect(globalProvider.models.find((model) => model.id === "glm-5.2")?.contextWindow).toBe(
1_000_000,
);
expect(cnProvider.models.find((model) => model.id === "glm-5.2")?.contextWindow).toBe(
1_000_000,
);
});
it("uses current model limits instead of the stale contributor catalog", () => {
const provider = buildQwenTokenPlanProvider();
expect(provider.models.find((model) => model.id === "qwen3.6-flash")?.maxTokens).toBe(65_536);
expect(provider.models.find((model) => model.id === "deepseek-v4-pro")).toMatchObject({
contextWindow: 1_000_000,
maxTokens: 393_216,
});
expect(provider.models.find((model) => model.id === "kimi-k2.7-code")?.maxTokens).toBe(98_304);
expect(provider.models.find((model) => model.id === "MiniMax-M2.5")).toMatchObject({
contextWindow: 196_608,
maxTokens: 32_768,
});
});
it.each([QWEN_TOKEN_PLAN_GLOBAL_BASE_URL, QWEN_TOKEN_PLAN_CN_BASE_URL])(
"opts Token Plan endpoint %s into native streaming usage",
(baseUrl) => {
const provider = applyQwenNativeStreamingUsageCompat(buildQwenTokenPlanProvider({ baseUrl }));
expect(provider.models).toHaveLength(14);
expect(
provider.models.every((model) => model.compat?.supportsUsageInStreaming === true),
).toBe(true);
},
);
});