Aggregate reasoning usage and derive cache hit ratios
This commit is contained in:
@@ -317,7 +317,7 @@ func TestRunnerRunPreparedUsesFrozenValidationForInitialAndRepairOutputs(t *test
|
||||
responses: []*domain.GenerateResponse{{
|
||||
Content: `{"repaired":true}`,
|
||||
Usage: domain.TokenUsage{
|
||||
PromptTokens: 2, CompletionTokens: 3, TotalTokens: 5,
|
||||
PromptTokens: 2, CompletionTokens: 3, ReasoningTokens: 2, TotalTokens: 5,
|
||||
CachedTokens: 7, CacheWriteTokens: 11,
|
||||
},
|
||||
}},
|
||||
@@ -328,7 +328,7 @@ func TestRunnerRunPreparedUsesFrozenValidationForInitialAndRepairOutputs(t *test
|
||||
client := &sequenceLLM{responses: []*domain.GenerateResponse{{
|
||||
Content: `{"broken":true}`,
|
||||
Usage: domain.TokenUsage{
|
||||
PromptTokens: 13, CompletionTokens: 17, TotalTokens: 19,
|
||||
PromptTokens: 13, CompletionTokens: 17, ReasoningTokens: 4, TotalTokens: 19,
|
||||
CachedTokens: 23, CacheWriteTokens: 29,
|
||||
},
|
||||
}}}
|
||||
@@ -370,7 +370,7 @@ func TestRunnerRunPreparedUsesFrozenValidationForInitialAndRepairOutputs(t *test
|
||||
t.Fatalf("unexpected repaired validation result: %+v", result.Validation)
|
||||
}
|
||||
wantUsage := domain.TokenUsage{
|
||||
PromptTokens: 15, CompletionTokens: 20, TotalTokens: 24,
|
||||
PromptTokens: 15, CompletionTokens: 20, ReasoningTokens: 6, TotalTokens: 24,
|
||||
CachedTokens: 30, CacheWriteTokens: 40,
|
||||
}
|
||||
if result.Usage != wantUsage {
|
||||
|
||||
@@ -309,6 +309,7 @@ func addTokenUsage(total, next domain.TokenUsage) domain.TokenUsage {
|
||||
return domain.TokenUsage{
|
||||
PromptTokens: total.PromptTokens + next.PromptTokens,
|
||||
CompletionTokens: total.CompletionTokens + next.CompletionTokens,
|
||||
ReasoningTokens: total.ReasoningTokens + next.ReasoningTokens,
|
||||
TotalTokens: total.TotalTokens + next.TotalTokens,
|
||||
CachedTokens: total.CachedTokens + next.CachedTokens,
|
||||
CacheWriteTokens: total.CacheWriteTokens + next.CacheWriteTokens,
|
||||
|
||||
Reference in New Issue
Block a user