getcompanion-ai/co-mono
1
0
Fork 0
mirror of https://github.com/getcompanion-ai/co-mono.git synced 2026-04-17 07:03:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat(ai): add cacheRetention stream option

Browse source
This commit is contained in:
Mario Zechner 2026-02-01 09:30:23 +01:00
parent e9ca0be769
commit abfd04b5c5
5 changed files with 174 additions and 39 deletions
Download patch file Download diff file Expand all files Collapse all files

108
packages/ai/test/cache-retention.test.ts
View file

@ -112,6 +112,58 @@ describe("Cache Retention (PI_CACHE_RETENTION)", () => {
expect(capturedPayload.system[0].cache_control).toEqual({ type: "ephemeral" });
}
});
it("should omit cache_control when cacheRetention is none", async () => {
const baseModel = getModel("anthropic", "claude-3-5-haiku-20241022");
let capturedPayload: any = null;
const { streamAnthropic } = await import("../src/providers/anthropic.js");
try {
const s = streamAnthropic(baseModel, context, {
apiKey: "fake-key",
cacheRetention: "none",
onPayload: (payload) => {
capturedPayload = payload;
},
});
for await (const event of s) {
if (event.type === "error") break;
}
} catch {
// Expected to fail
}
expect(capturedPayload).not.toBeNull();
expect(capturedPayload.system[0].cache_control).toBeUndefined();
});
it("should set 1h cache TTL when cacheRetention is long", async () => {
const baseModel = getModel("anthropic", "claude-3-5-haiku-20241022");
let capturedPayload: any = null;
const { streamAnthropic } = await import("../src/providers/anthropic.js");
try {
const s = streamAnthropic(baseModel, context, {
apiKey: "fake-key",
cacheRetention: "long",
onPayload: (payload) => {
capturedPayload = payload;
},
});
for await (const event of s) {
if (event.type === "error") break;
}
} catch {
// Expected to fail
}
expect(capturedPayload).not.toBeNull();
expect(capturedPayload.system[0].cache_control).toEqual({ type: "ephemeral", ttl: "1h" });
});
});
describe("OpenAI Responses Provider", () => {
@ -195,5 +247,61 @@ describe("Cache Retention (PI_CACHE_RETENTION)", () => {
expect(capturedPayload.prompt_cache_retention).toBeUndefined();
}
});
it("should omit prompt_cache_key when cacheRetention is none", async () => {
const model = getModel("openai", "gpt-4o-mini");
let capturedPayload: any = null;
const { streamOpenAIResponses } = await import("../src/providers/openai-responses.js");
try {
const s = streamOpenAIResponses(model, context, {
apiKey: "fake-key",
cacheRetention: "none",
sessionId: "session-1",
onPayload: (payload) => {
capturedPayload = payload;
},
});
for await (const event of s) {
if (event.type === "error") break;
}
} catch {
// Expected to fail
}
expect(capturedPayload).not.toBeNull();
expect(capturedPayload.prompt_cache_key).toBeUndefined();
expect(capturedPayload.prompt_cache_retention).toBeUndefined();
});
it("should set prompt_cache_retention when cacheRetention is long", async () => {
const model = getModel("openai", "gpt-4o-mini");
let capturedPayload: any = null;
const { streamOpenAIResponses } = await import("../src/providers/openai-responses.js");
try {
const s = streamOpenAIResponses(model, context, {
apiKey: "fake-key",
cacheRetention: "long",
sessionId: "session-2",
onPayload: (payload) => {
capturedPayload = payload;
},
});
for await (const event of s) {
if (event.type === "error") break;
}
} catch {
// Expected to fail
}
expect(capturedPayload).not.toBeNull();
expect(capturedPayload.prompt_cache_key).toBe("session-2");
expect(capturedPayload.prompt_cache_retention).toBe("24h");
});
});
});