diff --git a/backend/internal/service/billing_service.go b/backend/internal/service/billing_service.go
index e2046dd979..a781936598 100644
--- a/backend/internal/service/billing_service.go
+++ b/backend/internal/service/billing_service.go
@@ -276,8 +276,9 @@ func (s *BillingService) initFallbackPricing() {
 		LongContextInputMultiplier:     openAIGPT54LongContextInputMultiplier,
 		LongContextOutputMultiplier:    openAIGPT54LongContextOutputMultiplier,
 	}
-	// GPT-5.5 暂无独立定价，回退到 GPT-5.4
+	// GPT-5.5 / GPT-5.5 Pro 暂无独立定价，回退到 GPT-5.4。
 	s.fallbackPrices["gpt-5.5"] = s.fallbackPrices["gpt-5.4"]
+	s.fallbackPrices["gpt-5.5-pro"] = s.fallbackPrices["gpt-5.4"]
 
 	s.fallbackPrices["gpt-5.4-mini"] = &ModelPricing{
 		InputPricePerToken:     7.5e-7,
@@ -666,6 +667,8 @@ func (s *BillingService) getFallbackPricing(model string) *ModelPricing {
 	// OpenAI（GPT-5 / Codex 族）：仅匹配已知型号，避免未知 OpenAI 型号误计价。
 	if normalized := normalizeKnownOpenAICodexModel(modelLower); normalized != "" {
 		switch normalized {
+		case "gpt-5.5-pro":
+			return s.fallbackPrices["gpt-5.5-pro"]
 		case "gpt-5.5":
 			return s.fallbackPrices["gpt-5.5"]
 		case "gpt-5.4-mini":
@@ -1057,7 +1060,7 @@ func isOpenAIGPT54Model(model string) bool {
 	// normalizeCodexModel 的默认兜底把非 OpenAI 模型（claude-*、gemini-*、gpt-4o）
 	// 误识别为 gpt-5.4。
 	normalized := normalizeKnownOpenAICodexModel(model)
-	return normalized == "gpt-5.4" || normalized == "gpt-5.5"
+	return normalized == "gpt-5.4" || normalized == "gpt-5.5" || normalized == "gpt-5.5-pro"
 }
 
 // CalculateCostWithConfig 使用配置中的默认倍率计算费用
diff --git a/backend/internal/service/billing_service_test.go b/backend/internal/service/billing_service_test.go
index 62792dc5df..92c143c6ff 100644
--- a/backend/internal/service/billing_service_test.go
+++ b/backend/internal/service/billing_service_test.go
@@ -263,6 +263,25 @@ func TestCalculateCost_OpenAIGPT54LongContextAppliesWholeSessionMultipliers(t *t
 	require.InDelta(t, expectedInput+expectedOutput, cost.ActualCost, 1e-10)
 }
 
+func TestCalculateCost_OpenAIGPT55ProUsesGPT55PricingPolicy(t *testing.T) {
+	svc := newTestBillingService()
+
+	tokens := UsageTokens{
+		InputTokens:  300000,
+		OutputTokens: 4000,
+	}
+
+	cost, err := svc.CalculateCost("gpt-5.5-pro", tokens, 1.0)
+	require.NoError(t, err)
+
+	expectedInput := float64(tokens.InputTokens) * 2.5e-6 * 2.0
+	expectedOutput := float64(tokens.OutputTokens) * 15e-6 * 1.5
+	require.InDelta(t, expectedInput, cost.InputCost, 1e-10)
+	require.InDelta(t, expectedOutput, cost.OutputCost, 1e-10)
+	require.InDelta(t, expectedInput+expectedOutput, cost.TotalCost, 1e-10)
+	require.InDelta(t, expectedInput+expectedOutput, cost.ActualCost, 1e-10)
+}
+
 // 回归测试 #2293：长上下文计费触发时，cache_read_tokens 也应应用 LongContextInputMultiplier。
 // 修复前：CacheReadCost = tokens * 0.25e-6 （漏乘倍率，少计费用）。
 // 修复后：CacheReadCost = tokens * 0.25e-6 * LongContextInputMultiplier(=2.0)。
diff --git a/backend/internal/service/openai_codex_transform.go b/backend/internal/service/openai_codex_transform.go
index 4bb29167f9..c551ba54f1 100644
--- a/backend/internal/service/openai_codex_transform.go
+++ b/backend/internal/service/openai_codex_transform.go
@@ -10,6 +10,7 @@ import (
 
 var codexModelMap = map[string]string{
 	"gpt-5.5":                    "gpt-5.5",
+	"gpt-5.5-pro":                "gpt-5.5-pro",
 	"codex-auto-review":          "codex-auto-review",
 	"gpt-5.4":                    "gpt-5.4",
 	"gpt-5.4-mini":               "gpt-5.4-mini",
@@ -61,6 +62,7 @@ var codexVersionModelPrefixes = []struct {
 	{prefix: "gpt-5.3-codex", target: "gpt-5.3-codex"},
 	{prefix: "gpt-5.4-mini", target: "gpt-5.4-mini"},
 	{prefix: "gpt-5.4-nano", target: "gpt-5.4-nano"},
+	{prefix: "gpt-5.5-pro", target: "gpt-5.5-pro"},
 	{prefix: "gpt-5.5", target: "gpt-5.5"},
 	{prefix: "gpt-5.4", target: "gpt-5.4"},
 	{prefix: "gpt-5.2", target: "gpt-5.2"},
diff --git a/backend/internal/service/openai_codex_transform_test.go b/backend/internal/service/openai_codex_transform_test.go
index d3bdc0a52e..bb27c81352 100644
--- a/backend/internal/service/openai_codex_transform_test.go
+++ b/backend/internal/service/openai_codex_transform_test.go
@@ -898,6 +898,10 @@ func TestNormalizeCodexModel_Gpt53(t *testing.T) {
 		"gpt-5.4":                   "gpt-5.4",
 		"gpt5.5":                    "gpt-5.5",
 		"openai/gpt5.5":             "gpt-5.5",
+		"gpt-5.5-pro":               "gpt-5.5-pro",
+		"gpt5.5-pro":                "gpt-5.5-pro",
+		"openai/gpt5.5-pro":         "gpt-5.5-pro",
+		"gpt-5.5-pro-high":          "gpt-5.5-pro",
 		"codex-auto-review":         "codex-auto-review",
 		"gpt5.4":                    "gpt-5.4",
 		"gpt-5.4-high":              "gpt-5.4",
diff --git a/backend/internal/service/openai_compat_prompt_cache_key_test.go b/backend/internal/service/openai_compat_prompt_cache_key_test.go
index 3fe7db6ef6..ce1c68a2b7 100644
--- a/backend/internal/service/openai_compat_prompt_cache_key_test.go
+++ b/backend/internal/service/openai_compat_prompt_cache_key_test.go
@@ -16,6 +16,7 @@ func mustRawJSON(t *testing.T, s string) json.RawMessage {
 
 func TestShouldAutoInjectPromptCacheKeyForCompat(t *testing.T) {
 	require.True(t, shouldAutoInjectPromptCacheKeyForCompat("gpt-5.5"))
+	require.True(t, shouldAutoInjectPromptCacheKeyForCompat("gpt-5.5-pro"))
 	require.True(t, shouldAutoInjectPromptCacheKeyForCompat("gpt-5.4"))
 	require.True(t, shouldAutoInjectPromptCacheKeyForCompat("gpt-5.4-mini"))
 	require.True(t, shouldAutoInjectPromptCacheKeyForCompat("gpt-5.2"))
diff --git a/backend/internal/service/openai_model_alias.go b/backend/internal/service/openai_model_alias.go
index 2fa2c90efd..ac2a8cf942 100644
--- a/backend/internal/service/openai_model_alias.go
+++ b/backend/internal/service/openai_model_alias.go
@@ -65,6 +65,8 @@ func normalizeKnownOpenAICodexModel(model string) string {
 	}
 
 	switch {
+	case strings.Contains(normalized, "gpt-5.5-pro"):
+		return "gpt-5.5-pro"
 	case strings.Contains(normalized, "gpt-5.5"):
 		return "gpt-5.5"
 	case strings.Contains(normalized, "gpt-5.4-mini"):
diff --git a/backend/internal/service/openai_model_mapping_test.go b/backend/internal/service/openai_model_mapping_test.go
index 020e887528..0f65740be7 100644
--- a/backend/internal/service/openai_model_mapping_test.go
+++ b/backend/internal/service/openai_model_mapping_test.go
@@ -94,6 +94,15 @@ func TestResolveOpenAIForwardModel(t *testing.T) {
 			defaultMappedModel: "gpt-5.4",
 			expectedModel:      "gpt-5.5",
 		},
+		{
+			name: "preserves gpt-5.5-pro instead of group default",
+			account: &Account{
+				Credentials: map[string]any{},
+			},
+			requestedModel:     "gpt-5.5-pro",
+			defaultMappedModel: "gpt-5.5",
+			expectedModel:      "gpt-5.5-pro",
+		},
 		{
 			name: "preserves compact-spelled gpt5.5 instead of group default",
 			account: &Account{
@@ -261,6 +270,12 @@ func TestNormalizeOpenAIModelForUpstream(t *testing.T) {
 			model:   "gpt-5.4-high",
 			want:    "gpt-5.4",
 		},
+		{
+			name:    "oauth preserves GPT-5.5 Pro model",
+			account: &Account{Type: AccountTypeOAuth},
+			model:   "openai/gpt-5.5-pro",
+			want:    "gpt-5.5-pro",
+		},
 		{
 			name:    "oauth preserves codex auto review model",
 			account: &Account{Type: AccountTypeOAuth},
@@ -303,3 +318,17 @@ func TestUsageBillingModelCandidatesPreserveCodexAutoReviewModel(t *testing.T) {
 		}
 	}
 }
+
+func TestUsageBillingModelCandidatesPreserveGPT55ProModel(t *testing.T) {
+	candidates := usageBillingModelCandidates("openai/gpt-5.5-pro")
+
+	expected := []string{"openai/gpt-5.5-pro", "gpt-5.5-pro"}
+	if len(candidates) != len(expected) {
+		t.Fatalf("usageBillingModelCandidates(openai/gpt-5.5-pro) = %#v, want %#v", candidates, expected)
+	}
+	for i := range expected {
+		if candidates[i] != expected[i] {
+			t.Fatalf("usageBillingModelCandidates(openai/gpt-5.5-pro) = %#v, want %#v", candidates, expected)
+		}
+	}
+}