feat: sync deployed v1.0.0-rc.38-custom.67

service/detection.go

@@ -8,6 +8,7 @@ import (
 	"github.com/QuantumNous/new-api/common"
 	"github.com/QuantumNous/new-api/model"
 	"github.com/QuantumNous/new-api/pkg/billingexpr"
+	benchmark "github.com/QuantumNous/new-api/pkg/model_benchmark"
 	fingerprint "github.com/QuantumNous/new-api/pkg/model_fingerprint"
 	"github.com/QuantumNous/new-api/relaykit/dto"
 	"github.com/QuantumNous/new-api/setting/billing_setting"
@@ -36,15 +37,15 @@ func DetectionPricing(userGroup, group, kind string) (DetectionPrice, int, error
 		return DetectionPrice{}, 0, errors.New("invalid group ratio")
 	}
 	price := DetectionPrice{Snapshot: billingexpr.BillingSnapshot{BillingMode: "tiered_expr", ModelName: fingerprint.Model, ExprString: expression, ExprHash: billingexpr.ExprHashString(expression), GroupRatio: ratio, QuotaPerUnit: common.QuotaPerUnit, ExprVersion: 1}}
-	// Both generation paths restrict request text to 16 KiB and output tokens.
-	output, calls := 16384, 1
-	if kind == "fingerprint" {
-		output, calls = 4096, 3
-	}
-	price.Request.Body = []byte(fmt.Sprintf(`{"model":"gpt-6-astra","reasoning":{"effort":"high"},"max_output_tokens":%d}`, output))
-	if kind == "fingerprint" {
-		price.Request.Body = []byte(`{"model":"gpt-6-astra","max_output_tokens":4096}`)
+	output, calls := fingerprint.OutputTokens, fingerprint.ProbeCount
+	if kind == "benchmark" {
+		output, calls = benchmark.OutputTokens, benchmark.ProbeCount
+	} else if kind == "candy" {
+		output, calls = CandyOutputTokens, 1
+	} else if kind != "fingerprint" {
+		return price, 0, errors.New("invalid detection type")
 	}
+	price.Request.Body = []byte(fmt.Sprintf(`{"model":"gpt-6-astra","reasoning":{"effort":"low"},"max_output_tokens":%d}`, output))
 	cost, err := billingexpr.DetectionCostBound(expression, 32768, float64(output))
 	if err != nil {
 		return price, 0, err