feat(vertexai): use pointers for GenerationConfig fields

vertexai/genai/aiplatformpb_veneer.gen.go

@@ -479,15 +479,15 @@ func (GenerateContentResponse) fromProto(p *pb.GenerateContentResponse) *Generat
 // GenerationConfig is generation config.
 type GenerationConfig struct {
 	// Optional. Controls the randomness of predictions.
-	Temperature float32
+	Temperature *float32
 	// Optional. If specified, nucleus sampling will be used.
-	TopP float32
+	TopP *float32
 	// Optional. If specified, top-k sampling will be used.
-	TopK float32
+	TopK *float32
 	// Optional. Number of candidates to generate.
-	CandidateCount int32
+	CandidateCount *int32
 	// Optional. The maximum number of output tokens to generate per message.
-	MaxOutputTokens int32
+	MaxOutputTokens *int32
 	// Optional. Stop sequences.
 	StopSequences []string
 }
@@ -497,11 +497,11 @@ func (v *GenerationConfig) toProto() *pb.GenerationConfig {
 		return nil
 	}
 	return &pb.GenerationConfig{
-		Temperature:     support.AddrOrNil(v.Temperature),
-		TopP:            support.AddrOrNil(v.TopP),
-		TopK:            support.AddrOrNil(v.TopK),
-		CandidateCount:  support.AddrOrNil(v.CandidateCount),
-		MaxOutputTokens: support.AddrOrNil(v.MaxOutputTokens),
+		Temperature:     v.Temperature,
+		TopP:            v.TopP,
+		TopK:            v.TopK,
+		CandidateCount:  v.CandidateCount,
+		MaxOutputTokens: v.MaxOutputTokens,
 		StopSequences:   v.StopSequences,
 	}
 }
@@ -511,11 +511,11 @@ func (GenerationConfig) fromProto(p *pb.GenerationConfig) *GenerationConfig {
 		return nil
 	}
 	return &GenerationConfig{
-		Temperature:     support.DerefOrZero(p.Temperature),
-		TopP:            support.DerefOrZero(p.TopP),
-		TopK:            support.DerefOrZero(p.TopK),
-		CandidateCount:  support.DerefOrZero(p.CandidateCount),
-		MaxOutputTokens: support.DerefOrZero(p.MaxOutputTokens),
+		Temperature:     p.Temperature,
+		TopP:            p.TopP,
+		TopK:            p.TopK,
+		CandidateCount:  p.CandidateCount,
+		MaxOutputTokens: p.MaxOutputTokens,
 		StopSequences:   p.StopSequences,
 	}
 }

vertexai/genai/client.go

@@ -90,10 +90,6 @@ const defaultMaxOutputTokens = 2048
 // GenerativeModel creates a new instance of the named model.
 func (c *Client) GenerativeModel(name string) *GenerativeModel {
 	return &GenerativeModel{
-		GenerationConfig: GenerationConfig{
-			MaxOutputTokens: defaultMaxOutputTokens,
-			TopK:            3,
-		},
 		c:        c,
 		name:     name,
 		fullName: fmt.Sprintf("projects/%s/locations/%s/publishers/google/models/%s", c.projectID, c.location, name),

vertexai/genai/client_test.go

@@ -49,7 +49,7 @@ func TestLive(t *testing.T) {
 	}
 	defer client.Close()
 	model := client.GenerativeModel(*modelName)
-	model.Temperature = 0
+	model.Temperature = Ptr[float32](0)
 
 	t.Run("GenerateContent", func(t *testing.T) {
 		resp, err := model.GenerateContent(ctx, Text("What is the average size of a swallow?"))
@@ -104,16 +104,16 @@ func TestLive(t *testing.T) {
 
 		checkMatch(t,
 			send("Which is best?", true),
-			"best", "air fryer", "Philips", "([Cc]onsider|research|compare)", "factors|features")
+			"best", "air fryer", "Philips", "([Cc]onsider|research|compare|preference)", "factors|features")
 
 		checkMatch(t,
 			send("Say that again.", false),
-			"best", "air fryer", "Philips", "([Cc]onsider|research|compare)", "factors|features")
+			"best", "air fryer", "Philips", "([Cc]onsider|research|compare|preference)", "factors|features")
 	})
 
 	t.Run("image", func(t *testing.T) {
 		vmodel := client.GenerativeModel(*modelName + "-vision")
-		vmodel.Temperature = 0
+		vmodel.Temperature = Ptr[float32](0)
 
 		data, err := os.ReadFile(filepath.Join("testdata", imageFile))
 		if err != nil {
@@ -168,8 +168,8 @@ func TestLive(t *testing.T) {
 	})
 	t.Run("max-tokens", func(t *testing.T) {
 		maxModel := client.GenerativeModel(*modelName)
-		maxModel.Temperature = 0
-		maxModel.MaxOutputTokens = 10
+		maxModel.Temperature = Ptr(float32(0))
+		maxModel.SetMaxOutputTokens(10)
 		res, err := maxModel.GenerateContent(ctx, Text("What is a dog?"))
 		if err != nil {
 			t.Fatal(err)
@@ -182,8 +182,8 @@ func TestLive(t *testing.T) {
 	})
 	t.Run("max-tokens-streaming", func(t *testing.T) {
 		maxModel := client.GenerativeModel(*modelName)
-		maxModel.Temperature = 0
-		maxModel.MaxOutputTokens = 10
+		maxModel.Temperature = Ptr[float32](0)
+		maxModel.MaxOutputTokens = Ptr[int32](10)
 		iter := maxModel.GenerateContentStream(ctx, Text("What is a dog?"))
 		var merged *GenerateContentResponse
 		for {
@@ -232,7 +232,7 @@ func TestLive(t *testing.T) {
 			}},
 		}
 		model := client.GenerativeModel(*modelName)
-		model.Temperature = 0
+		model.SetTemperature(0)
 		model.Tools = []*Tool{weatherTool}
 		session := model.StartChat()
 		res, err := session.SendMessage(ctx, Text("What is the weather like in New York?"))
@@ -470,3 +470,19 @@ func TestMatchString(t *testing.T) {
 		}
 	}
 }
+
+func TestTemperature(t *testing.T) {
+	m := &GenerativeModel{}
+	got := m.GenerationConfig.toProto().Temperature
+	if got != nil {
+		t.Errorf("got %v, want nil", got)
+	}
+	m.SetTemperature(0)
+	got = m.GenerationConfig.toProto().Temperature
+	if got == nil {
+		t.Fatal("got nil")
+	}
+	if g := *got; g != 0 {
+		t.Errorf("got %v, want 0", g)
+	}
+}

vertexai/genai/config.yaml

@@ -59,17 +59,6 @@ types:
     FunctionResponse:
 
     GenerationConfig:
-      fields:
-        Temperature:
-          type: float32
-        TopP:
-          type: float32
-        TopK:
-          type: float32
-        CandidateCount:
-          type: int32
-        MaxOutputTokens:
-          type: int32
 
     SafetyRating:
       docVerb: 'is the'

vertexai/genai/content.go

@@ -113,3 +113,24 @@ func ImageData(format string, data []byte) Blob {
 		Data:     data,
 	}
 }
+
+// Ptr returns a pointer to its argument.
+// It can be used to initialize pointer fields:
+//
+//	model.Temperature = genai.Ptr[float32](0.1)
+func Ptr[T any](t T) *T { return &t }
+
+// SetCandidateCount sets the CandidateCount field.
+func (c *GenerationConfig) SetCandidateCount(x int32) { c.CandidateCount = &x }
+
+// SetMaxOutputTokens sets the MaxOutputTokens field.
+func (c *GenerationConfig) SetMaxOutputTokens(x int32) { c.MaxOutputTokens = &x }
+
+// SetTemperature sets the Temperature field.
+func (c *GenerationConfig) SetTemperature(x float32) { c.Temperature = &x }
+
+// SetTopP sets the TopP field.
+func (c *GenerationConfig) SetTopP(x float32) { c.TopP = &x }
+
+// SetTopK sets the TopK field.
+func (c *GenerationConfig) SetTopK(x float32) { c.TopK = &x }

vertexai/genai/example_test.go

@@ -42,7 +42,7 @@ func ExampleGenerativeModel_GenerateContent() {
 	defer client.Close()
 
 	model := client.GenerativeModel(model)
-	model.Temperature = 0.9
+	model.SetTemperature(0.9)
 	resp, err := model.GenerateContent(ctx, genai.Text("What is the average size of a swallow?"))
 	if err != nil {
 		log.Fatal(err)