diff --git a/config.go b/config.go
index e5f5ce9..3802c71 100644
--- a/config.go
+++ b/config.go
@@ -17,37 +17,34 @@ import (
 
 const configTemplate = `
 # {{ index .Help "apis" }}
+# LocalAI setup instructions: https://github.com/go-skynet/LocalAI#example-use-gpt4all-j-model
 apis:
-  openai: https://api.openai.com/v1
-  # LocalAI setup instructions: https://github.com/go-skynet/LocalAI#example-use-gpt4all-j-model
-  localai: http://localhost:8080
-# {{ index .Help "models" }}
-models:
-  gpt-4:
-    aliases: ["4"]
-    max-input-chars: 24500
-    api: openai
-    fallback: gpt-3.5-turbo
-  gpt-4-32k:
-    aliases: ["32k"]
-    max-input-chars: 98000
-    api: openai
-    fallback: gpt-4
-  gpt-3.5-turbo:
-    aliases: ["35t"]
-    max-input-chars: 12250
-    api: openai
-    fallback: gpt-3.5
-  gpt-3.5:
-    aliases: ["35"]
-    max-input-chars: 12250
-    api: openai
-    fallback:
-  ggml-gpt4all-j:
-    aliases: ["local", "4all"]
-    max-input-chars: 12250
-    api: localai
-    fallback:
+  openai:
+    base-url: https://api.openai.com/v1
+    models:
+      gpt-4:
+        aliases: ["4"]
+        max-input-chars: 24500
+        fallback: gpt-3.5-turbo
+      gpt-4-32k:
+        aliases: ["32k"]
+        max-input-chars: 98000
+        fallback: gpt-4
+      gpt-3.5-turbo:
+        aliases: ["35t"]
+        max-input-chars: 12250
+        fallback: gpt-3.5
+      gpt-3.5:
+        aliases: ["35"]
+        max-input-chars: 12250
+        fallback:
+  localai:
+    base-url: http://localhost:8080
+    models:
+      ggml-gpt4all-j:
+        aliases: ["local", "4all"]
+        max-input-chars: 12250
+        fallback:
 # {{ index .Help "model" }}
 default-model: gpt-4
 # {{ index .Help "max-input-chars" }}
@@ -77,22 +74,22 @@ status-text: Generating
 `
 
 type config struct {
-	API               string            `yaml:"api" env:"API"`
-	APIs              map[string]string `yaml:"apis"`
-	Model             string            `yaml:"default-model" env:"MODEL"`
-	Models            map[string]Model  `yaml:"models"`
-	Markdown          bool              `yaml:"format" env:"FORMAT"`
-	Quiet             bool              `yaml:"quiet" env:"QUIET"`
-	MaxTokens         int               `yaml:"max-tokens" env:"MAX_TOKENS"`
-	MaxInputChars     int               `yaml:"max-input-chars" env:"MAX_INPUT_CHARS"`
-	Temperature       float32           `yaml:"temp" env:"TEMP"`
-	TopP              float32           `yaml:"topp" env:"TOPP"`
-	NoLimit           bool              `yaml:"no-limit" env:"NO_LIMIT"`
-	IncludePromptArgs bool              `yaml:"include-prompt-args" env:"INCLUDE_PROMPT_ARGS"`
-	IncludePrompt     int               `yaml:"include-prompt" env:"INCLUDE_PROMPT"`
-	MaxRetries        int               `yaml:"max-retries" env:"MAX_RETRIES"`
-	Fanciness         uint              `yaml:"fanciness" env:"FANCINESS"`
-	StatusText        string            `yaml:"status-text" env:"STATUS_TEXT"`
+	APIs              map[string]API `yaml:"apis"`
+	Model             string         `yaml:"default-model" env:"MODEL"`
+	Markdown          bool           `yaml:"format" env:"FORMAT"`
+	Quiet             bool           `yaml:"quiet" env:"QUIET"`
+	MaxTokens         int            `yaml:"max-tokens" env:"MAX_TOKENS"`
+	MaxInputChars     int            `yaml:"max-input-chars" env:"MAX_INPUT_CHARS"`
+	Temperature       float32        `yaml:"temp" env:"TEMP"`
+	TopP              float32        `yaml:"topp" env:"TOPP"`
+	NoLimit           bool           `yaml:"no-limit" env:"NO_LIMIT"`
+	IncludePromptArgs bool           `yaml:"include-prompt-args" env:"INCLUDE_PROMPT_ARGS"`
+	IncludePrompt     int            `yaml:"include-prompt" env:"INCLUDE_PROMPT"`
+	MaxRetries        int            `yaml:"max-retries" env:"MAX_RETRIES"`
+	Fanciness         uint           `yaml:"fanciness" env:"FANCINESS"`
+	StatusText        string         `yaml:"status-text" env:"STATUS_TEXT"`
+	API               string
+	Models            map[string]Model
 	ShowHelp          bool
 	Prefix            string
 	Version           bool
@@ -105,10 +102,9 @@ func newConfig() (config, error) {
 	var content []byte
 
 	help := map[string]string{
-		"api":             "Default OpenAI compatible REST API (openai, localai).",
+		"api":             "OpenAI compatible REST API (openai, localai).",
 		"apis":            "Aliases and endpoints for OpenAI compatible REST API.",
 		"model":           "Default model (gpt-3.5-turbo, gpt-4, ggml-gpt4all-j...).",
-		"models":          "Model details and aliases.",
 		"max-input-chars": "Default character limit on input to model.",
 		"format":          "Format response as markdown.",
 		"prompt":          "Include the prompt from the arguments and stdin, truncate stdin to specified number of lines.",
@@ -168,13 +164,15 @@ func newConfig() (config, error) {
 		return c, err
 	}
 
-	// Set model aliases
 	ms := make(map[string]Model)
-	for k, m := range c.Models {
-		m.Name = k
-		ms[k] = m
-		for _, am := range m.Aliases {
-			ms[am] = m
+	for ak, av := range c.APIs {
+		for mk, mv := range av.Models {
+			mv.Name = mk
+			mv.API = ak
+			ms[mk] = mv
+			for _, a := range mv.Aliases {
+				ms[a] = mv
+			}
 		}
 	}
 	c.Models = ms
diff --git a/model.go b/model.go
index e0f18b7..2c7a993 100644
--- a/model.go
+++ b/model.go
@@ -3,8 +3,14 @@ package main
 // Model represents the LLM model used in the API call.
 type Model struct {
 	Name     string
+	API      string
 	MaxChars int      `yaml:"max-input-chars"`
 	Aliases  []string `yaml:"aliases"`
-	API      string   `yaml:"api"`
 	Fallback string   `yaml:"fallback"`
 }
+
+// API represents an API endpoint and its models.
+type API struct {
+	BaseURL string           `yaml:"base-url"`
+	Models  map[string]Model `yaml:"models"`
+}
diff --git a/mods.go b/mods.go
index cc913af..a42426d 100644
--- a/mods.go
+++ b/mods.go
@@ -226,7 +226,7 @@ func (m *Mods) startCompletionCmd(content string) tea.Cmd {
 			}
 		}
 		ccfg := openai.DefaultConfig(key)
-		ccfg.BaseURL, ok = cfg.APIs[mod.API]
+		api, ok := cfg.APIs[mod.API]
 		if !ok {
 			eps := make([]string, 0)
 			for k := range cfg.APIs {
@@ -237,6 +237,7 @@ func (m *Mods) startCompletionCmd(content string) tea.Cmd {
 				err:    fmt.Errorf("Your configured API endpoints are: %s", eps),
 			}
 		}
+		ccfg.BaseURL = api.BaseURL
 		client := openai.NewClientWithConfig(ccfg)
 		ctx, cancel := context.WithCancel(context.Background())
 		defer cancel()