The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM]
Python ai-gatewayanthropicazure-openaibedrockgatewaylangchainlitellmllmllm-gatewayllmopsmcp-gatewayopenaiopenai-proxyrustrust-aivertex-ai
import { describe, expect, it } from "vitest";
import type { components } from "@/lib/http/schema";
import {
EMPTY_DAILY_ACTIVITY_METADATA,
type DailyActivityAggregatedResponse,
type KeySpendActivityRow,
toDailyData,
} from "./dailyActivityApi";
import { keyDetailFromResponse, mergeKeyActivityPages, overallUsageMetrics } from "./keyActivityData";
const completeMetrics: components["schemas"]["SpendMetrics"] = {
api_requests: 2,
autorouter_savings_spend: 0,
cache_creation_input_tokens: 0,
cache_read_input_tokens: 0,
completion_tokens: 3,
compression_saved_tokens: 0,
compression_savings_spend: 0,
failed_requests: 0,
flat_cost: 0,
gateway_injected_caching_savings_spend: 0,
prompt_caching_savings_spend: 0,
prompt_tokens: 4,
spend: 1.25,
successful_requests: 2,
timed_requests: 0,
total_response_time_ms: 0,
total_tokens: 7,
};
const apiKeyActivity = {
metrics: completeMetrics,
metadata: { key_alias: "visible-key", team_id: null },
};
const aggregatedResponse: DailyActivityAggregatedResponse = {
metadata: {
...EMPTY_DAILY_ACTIVITY_METADATA,
total_api_requests: 200,
total_successful_requests: 198,
total_failed_requests: 2,
total_tokens: 700,
total_prompt_tokens: 400,
total_completion_tokens: 300,
total_spend: 500,
},
results: [
{
date: "2026-09-27",
metrics: completeMetrics,
breakdown: {
api_keys: { "key-hash": apiKeyActivity },
models: {
"gpt-4o-mini": {
metrics: completeMetrics,
metadata: {},
api_key_breakdown: { "key-hash": apiKeyActivity },
},
},
},
},
],
};
const pageRow = (api_key: string): KeySpendActivityRow => ({
api_key,
metrics: {
api_requests: 1,
cache_creation_input_tokens: 0,
cache_read_input_tokens: 0,
completion_tokens: 2,
failed_requests: 0,
prompt_tokens: 3,
spend: 1,
successful_requests: 1,
total_tokens: 5,
},
metadata: { key_alias: api_key, team_id: null },
});
describe("key activity data", () => {
it("builds overall totals from metadata and daily data from complete daily metrics", () => {
const summary = overallUsageMetrics(
toDailyData(aggregatedResponse),
aggregatedResponse.metadata ?? EMPTY_DAILY_ACTIVITY_METADATA,
);
expect(summary.total_requests).toBe(200);
expect(summary.total_spend).toBe(500);
expect(summary.daily_data).toStrictEqual([
{
date: "2026-09-27",
metrics: {
prompt_tokens: 4,
completion_tokens: 3,
total_tokens: 7,
api_requests: 2,
spend: 1.25,
successful_requests: 2,
failed_requests: 0,
cache_read_input_tokens: 0,
cache_creation_input_tokens: 0,
avg_response_time_ms: null,
},
},
]);
});
it("appends pages without duplicate keys and compares the server offset to the total", () => {
const merged = mergeKeyActivityPages(
[pageRow("key-a"), pageRow("key-b")],
[pageRow("key-b"), pageRow("key-c")],
5,
2,
);
expect(merged.rows.map((row) => row.api_key)).toStrictEqual(["key-a", "key-b", "key-c"]);
expect(merged.nextOffset).toBe(4);
expect(merged.hasMore).toBe(true);
expect(mergeKeyActivityPages(merged.rows, [pageRow("key-d")], 5, 4).hasMore).toBe(false);
});
it("advances past a page of already loaded keys and stops only on an empty page", () => {
const current = [pageRow("key-a"), pageRow("key-b")];
const duplicates = mergeKeyActivityPages(current, [pageRow("key-a"), pageRow("key-b")], 5, 2);
expect(duplicates).toEqual({ rows: current, nextOffset: 4, hasMore: true });
expect(mergeKeyActivityPages(current, [], 5, 2)).toEqual({ rows: current, nextOffset: 2, hasMore: false });
});
it("builds full daily details and top models for the requested key", () => {
const detail = keyDetailFromResponse(aggregatedResponse, "key-hash", []);
expect(detail?.total_requests).toBe(2);
expect(detail?.daily_data).toStrictEqual([
{
date: "2026-09-27",
metrics: {
prompt_tokens: 4,
completion_tokens: 3,
total_tokens: 7,
api_requests: 2,
spend: 1.25,
successful_requests: 2,
failed_requests: 0,
cache_read_input_tokens: 0,
cache_creation_input_tokens: 0,
avg_response_time_ms: null,
},
},
]);
expect(detail?.top_models.map((model) => model.model)).toStrictEqual(["gpt-4o-mini"]);
});
});