From de119ea47f9b51a02aafaa4499ef04584288fc47 Mon Sep 17 00:00:00 2001
From: Dong Chen <dongluochen@ollama.com>
Date: Wed, 30 Sep 2026 13:02:36 -0700
Subject: [PATCH] openai: keep tool message content parts in one message

FromChatRequest splits array content into one api.Message per part. That
is intended for user content, but a tool message is a single result: the
split also dropped its tool_call_id and tool name, since those were only
copied when the message had tool_calls. Renderers that resolve tool
results by ID then see extra, unnamed results; for example, Kimi K3 fails
with "tool message 2 has no resolvable tool name" on a tool result with
two text parts, which the OpenAI spec allows.

Merge the parts of a tool message into one message with the text joined,
the images kept, and ToolCallID and ToolName set, matching how
function_call_output items are converted for /v1/responses.
---
 openai/openai.go      | 11 +++++++++
 openai/openai_test.go | 54 +++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 65 insertions(+)

diff --git a/openai/openai.go b/openai/openai.go
index 010481518..efed4e048 100644
--- a/openai/openai.go
+++ b/openai/openai.go
@@ -634,6 +634,7 @@ func FromChatRequest(r ChatCompletionRequest, thinking ...*model.Thinking) (*api
 			}
 			messages = append(messages, api.Message{Role: msg.Role, Content: content, Thinking: msg.Reasoning, ToolCalls: toolCalls, ToolName: toolName, ToolCallID: msg.ToolCallID})
 		case []any:
+			start := len(messages)
 			for _, c := range content {
 				data, ok := c.(map[string]any)
 				if !ok {
@@ -682,6 +683,16 @@ func FromChatRequest(r ChatCompletionRequest, thinking ...*model.Thinking) (*api
 					return nil, errors.New("invalid message format")
 				}
 			}
+			// a tool message is a single result, so keep its parts together and
+			// preserve the tool call it answers
+			if strings.ToLower(msg.Role) == "tool" && len(messages) > start {
+				tool := api.Message{Role: msg.Role, ToolName: toolName, ToolCallID: msg.ToolCallID}
+				for _, part := range messages[start:] {
+					tool.Content += part.Content
+					tool.Images = append(tool.Images, part.Images...)
+				}
+				messages = append(messages[:start], tool)
+			}
 			// since we might have added multiple messages above, if we have tools
 			// calls we'll add them to the last message
 			if len(messages) > 0 && len(msg.ToolCalls) > 0 {
diff --git a/openai/openai_test.go b/openai/openai_test.go
index c92e19563..9f32d1df8 100644
--- a/openai/openai_test.go
+++ b/openai/openai_test.go
@@ -157,6 +157,60 @@ func TestFromChatRequest_WithImage(t *testing.T) {
 	}
 }
 
+func TestFromChatRequest_ToolMessageArrayContent(t *testing.T) {
+	imgData, _ := base64.StdEncoding.DecodeString(image)
+
+	req := ChatCompletionRequest{
+		Model: "test-model",
+		Messages: []Message{
+			{Role: "user", Content: "What's in the screenshot?"},
+			{
+				Role: "assistant",
+				ToolCalls: []ToolCall{{
+					ID:   "call_1",
+					Type: "function",
+					Function: struct {
+						Name      string `json:"name"`
+						Arguments string `json:"arguments"`
+					}{Name: "screenshot", Arguments: "{}"},
+				}},
+			},
+			{
+				Role:       "tool",
+				ToolCallID: "call_1",
+				Content: []any{
+					map[string]any{"type": "text", "text": "Captured "},
+					map[string]any{
+						"type":      "image_url",
+						"image_url": map[string]any{"url": prefix + image},
+					},
+					map[string]any{"type": "text", "text": "screen.png"},
+				},
+			},
+		},
+	}
+
+	result, err := FromChatRequest(req)
+	if err != nil {
+		t.Fatalf("unexpected error: %v", err)
+	}
+
+	if len(result.Messages) != 3 {
+		t.Fatalf("expected 3 messages, got %d", len(result.Messages))
+	}
+
+	tool := result.Messages[2]
+	if tool.Role != "tool" || tool.ToolCallID != "call_1" || tool.ToolName != "screenshot" {
+		t.Errorf("expected tool message for call_1/screenshot, got role=%q id=%q name=%q", tool.Role, tool.ToolCallID, tool.ToolName)
+	}
+	if tool.Content != "Captured screen.png" {
+		t.Errorf("expected content 'Captured screen.png', got %q", tool.Content)
+	}
+	if len(tool.Images) != 1 || string(tool.Images[0]) != string(imgData) {
+		t.Errorf("expected the tool image to be kept, got %d images", len(tool.Images))
+	}
+}
+
 func TestFromCompleteRequest_Basic(t *testing.T) {
 	temp := float32(0.8)
 	req := CompletionRequest{