feat: add Gemini 2.5 model support and update defaults

Replace older Gemini 2.0 models with the latest 2.5 series
(gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite).
Default model changed to gemini-2.5-flash for better out-of-box
performance. Bump version to 1.1.0.

Author: agnath18K

pkg/ai/gemini.go

@@ -54,7 +54,7 @@ type GeminiError struct {
 func NewGeminiClient(apiKey string, model string) *GeminiClient {
 	// If model is empty, use a default model
 	if model == "" {
-		model = "gemini-2.0-flash-lite"
+		model = "gemini-2.5-flash"
 	}
 
 	return &GeminiClient{

pkg/config/config.go

@@ -71,7 +71,7 @@ func DefaultConfig() *Config {
 	return &Config{
 		AIProvider:                  "gemini",                 // Default to Gemini
 		GeminiAPIKey:                "",                       // Will be loaded from environment
-		GeminiModel:                 "gemini-2.0-flash-lite",  // Default Gemini model
+		GeminiModel:                 "gemini-2.5-flash",       // Default Gemini model
 		OpenAIAPIKey:                "",                       // Will be loaded from environment
 		OpenAIModel:                 "gpt-3.5-turbo",          // Default OpenAI model
 		OllamaURL:                   "http://localhost:11434", // Default Ollama URL

pkg/executor/config.go

@@ -285,7 +285,7 @@ func (e *Executor) handleModelConfig(args []string, cmd *nlp.Command) (*Result,
 		// Validate model based on provider
 		switch e.config.AIProvider {
 		case "gemini":
-			validModels := []string{"gemini-2.0-flash-lite", "gemini-2.0-flash", "gemini-2.0-pro"}
+			validModels := []string{"gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite"}
 			isValid := false
 			for _, validModel := range validModels {
 				if model == validModel {

pkg/executor/model_list.go

@@ -15,9 +15,9 @@ func (e *Executor) handleModelList(cmd *nlp.Command) (*Result, error) {
 		output = `
 ╭─────────────── 🐦 Available Gemini Models ───────────────╮
 
-  • gemini-2.0-flash-lite  (Fast, efficient for most queries)
-  • gemini-2.0-flash       (Balanced performance and quality)
-  • gemini-2.0-pro         (High quality, more capabilities)
+  • gemini-2.5-pro         (Most capable, complex tasks)
+  • gemini-2.5-flash       (Fast, great balance of speed/quality)
+  • gemini-2.5-flash-lite  (Lightweight, fastest in 2.5 series)
 
   Current model: ` + e.config.GeminiModel + `
 

pkg/version/version.go

@@ -8,10 +8,10 @@ import (
 // Version information
 var (
 	// Version is the current version of Lumo
-	Version = "1.0.2"
+	Version = "1.1.0"
 
 	// BuildDate is the date when the binary was built
-	BuildDate = "May 19 2025"
+	BuildDate = "Mar 16 2026"
 
 	// GitCommit is the git commit hash when the binary was built
 	GitCommit = "HEAD"

tests/config_test.go

@@ -16,8 +16,8 @@ func TestConfigDefaultValues(t *testing.T) {
 		t.Errorf("Expected default AIProvider to be 'gemini', got '%s'", cfg.AIProvider)
 	}
 
-	if cfg.GeminiModel != "gemini-2.0-flash-lite" {
-		t.Errorf("Expected default GeminiModel to be 'gemini-2.0-flash-lite', got '%s'", cfg.GeminiModel)
+	if cfg.GeminiModel != "gemini-2.5-flash" {
+		t.Errorf("Expected default GeminiModel to be 'gemini-2.5-flash', got '%s'", cfg.GeminiModel)
 	}
 
 	if cfg.OpenAIModel != "gpt-3.5-turbo" {