test(stream): assert usage input excludes cached/written tokens
Follow the fix in stream.ts: prompt_tokens (42) includes cached_tokens (5), so input = 42 - 5 - 10 = 27. Use cache_write_tokens (not cacheable_tokens) for cacheWrite, and keep cacheable_tokens:99 in the payload to prove it is no longer mapped to cacheWrite.
This commit is contained in:
@@ -161,7 +161,12 @@ describe("streamQoder", () => {
|
||||
completion_tokens: 7,
|
||||
total_tokens: 49,
|
||||
completion_tokens_details: { reasoning_tokens: 3 },
|
||||
prompt_tokens_details: { cacheable_tokens: 10, cached_tokens: 5 },
|
||||
// prompt_tokens (42) INCLUDES cached_tokens (5) per OpenAI
|
||||
// semantics; pi-core expects `input` to exclude them
|
||||
// (promptTokens = input + cacheRead + cacheWrite), so input =
|
||||
// 42 - 5 - 10 = 27. cacheable_tokens is a capacity metric, not a
|
||||
// write count, and must not be mapped to cacheWrite.
|
||||
prompt_tokens_details: { cacheable_tokens: 99, cache_write_tokens: 10, cached_tokens: 5 },
|
||||
},
|
||||
}),
|
||||
) +
|
||||
@@ -174,7 +179,7 @@ describe("streamQoder", () => {
|
||||
const msg = (done as { message: AssistantMessage }).message;
|
||||
expect(msg.responseId).toBe("chatcmpl-abc123");
|
||||
expect(msg.responseModel).toBe("qmodel_latest");
|
||||
expect(msg.usage.input).toBe(42);
|
||||
expect(msg.usage.input).toBe(27);
|
||||
expect(msg.usage.output).toBe(7);
|
||||
expect(msg.usage.totalTokens).toBe(49);
|
||||
expect(msg.usage.cacheRead).toBe(5);
|
||||
|
||||
Reference in New Issue
Block a user