diff --git a/ldai/judge/judge.go b/ldai/judge/judge.go
index 608dd67a..1cad7d5d 100644
--- a/ldai/judge/judge.go
+++ b/ldai/judge/judge.go
@@ -21,7 +21,7 @@ type Config interface {
 
 type Tracker interface {
 	TrackJudgeResponse(response datamodel.JudgeResponse) error
-	TrackUsage(usage ldai.TokenUsage) error
+	TrackTokens(usage ldai.TokenUsage) error
 }
 
 type StructuredResponse struct {
@@ -91,7 +91,7 @@ func (j *Judge) Evaluate(input, output string, samplingRate float64) (*datamodel
 	}
 
 	if response.Usage.Total > 0 || response.Usage.Input > 0 || response.Usage.Output > 0 {
-		_ = j.tracker.TrackUsage(response.Usage)
+		_ = j.tracker.TrackTokens(response.Usage)
 	}
 
 	result := j.parseResponse(response.Content)
diff --git a/ldai/judge/judge_test.go b/ldai/judge/judge_test.go
index df0fd451..da0d7e55 100644
--- a/ldai/judge/judge_test.go
+++ b/ldai/judge/judge_test.go
@@ -51,7 +51,7 @@ func (m *mockTracker) TrackJudgeResponse(response datamodel.JudgeResponse) error
 	return nil
 }
 
-func (m *mockTracker) TrackUsage(usage ldai.TokenUsage) error {
+func (m *mockTracker) TrackTokens(usage ldai.TokenUsage) error {
 	m.usages = append(m.usages, usage)
 	return nil
 }
diff --git a/ldai/tracker.go b/ldai/tracker.go
index 90655242..ff380ccb 100644
--- a/ldai/tracker.go
+++ b/ldai/tracker.go
@@ -235,8 +235,8 @@ func (t *Tracker) TrackTimeToFirstToken(dur time.Duration) error {
 	return t.events.TrackMetric(timeToFirstToken, t.context, float64(dur.Milliseconds()), t.trackData)
 }
 
-// TrackUsage tracks the token usage for a model evaluation.
-func (t *Tracker) TrackUsage(usage TokenUsage) error {
+// TrackTokens tracks the token usage for a model evaluation.
+func (t *Tracker) TrackTokens(usage TokenUsage) error {
 	if usage.Set() {
 		t.tokens = ldcommon.Some(usage)
 	}
@@ -269,6 +269,13 @@ func (t *Tracker) TrackUsage(usage TokenUsage) error {
 	return nil
 }
 
+// TrackUsage tracks token usage.
+//
+// Deprecated: Use TrackTokens instead.
+func (t *Tracker) TrackUsage(usage TokenUsage) error {
+	return t.TrackTokens(usage)
+}
+
 func measureDurationOfTask[T any, A any](
 	stopwatch Stopwatch,
 	arg A,
@@ -335,8 +342,8 @@ func (t *Tracker) TrackRequest(task func(c *Config) (ProviderResponse, error)) (
 	}
 
 	if usage.Usage.Set() {
-		// TrackUsage logs errors.
-		_ = t.TrackUsage(usage.Usage)
+		// TrackTokens logs errors.
+		_ = t.TrackTokens(usage.Usage)
 	}
 
 	return usage, nil
diff --git a/ldai/tracker_test.go b/ldai/tracker_test.go
index 29bb71e4..836ff026 100644
--- a/ldai/tracker_test.go
+++ b/ldai/tracker_test.go
@@ -245,13 +245,13 @@ func TestTracker_TrackFeedback(t *testing.T) {
 	assert.ElementsMatch(t, []trackEvent{expectedPositiveEvent, expectedNegativeEvent}, events.events)
 }
 
-func TestTracker_TrackUsage(t *testing.T) {
+func TestTracker_TrackTokens(t *testing.T) {
 	t.Run("only one field set, only one event", func(t *testing.T) {
 		events := newMockEvents()
 		config := &Config{}
 		tracker := newTracker("key", "variationKey", 8, events, config, ldcontext.New("key"), nil)
 
-		assert.NoError(t, tracker.TrackUsage(TokenUsage{
+		assert.NoError(t, tracker.TrackTokens(TokenUsage{
 			Total: 42,
 		}))
 
@@ -270,7 +270,7 @@ func TestTracker_TrackUsage(t *testing.T) {
 		config := &Config{}
 		tracker := newTracker("key", "variationKey", 9, events, config, ldcontext.New("key"), nil)
 
-		assert.NoError(t, tracker.TrackUsage(TokenUsage{
+		assert.NoError(t, tracker.TrackTokens(TokenUsage{
 			Total:  42,
 			Input:  20,
 			Output: 22,
@@ -382,7 +382,7 @@ func TestTracker_GetSummary(t *testing.T) {
 			Input:  40,
 			Output: 60,
 		}
-		_ = tracker.TrackUsage(usage)
+		_ = tracker.TrackTokens(usage)
 
 		summary := tracker.GetSummary()