Added possibility to change MAX_TOKEN and also possible to set a session token that overrides MAX_TOKEN. None of them can exceed 1000000.

oai.py

@@ -14,7 +14,7 @@ import pyperclip
 import mimetypes
 import base64
 import re
-import sqlite3  # Added for SQLite DB integration
+import sqlite3
 from prompt_toolkit import PromptSession
 
 app = typer.Typer()
@@ -57,6 +57,7 @@ API_KEY = get_config('api_key')
 OPENROUTER_BASE_URL = get_config('base_url') or "https://openrouter.ai/api/v1"  # Default if not set
 STREAM_ENABLED = get_config('stream_enabled') or "on"  # Default to streaming on
 DEFAULT_MODEL_ID = get_config('default_model')  # Load default model ID from DB
+MAX_TOKEN = int(get_config('max_token') or "100000")  # Load max token limit, default to 100k
 
 # Fetch models once at module level (with loaded BASE_URL)
 models_data = []
@@ -122,8 +123,8 @@ def clear_screen():
 @app.command()
 def chat():
     """Start the oAI chat app with OpenRouter models."""
-    global API_KEY, OPENROUTER_BASE_URL, STREAM_ENABLED  # Allow updates via /config commands
-    
+    global API_KEY, OPENROUTER_BASE_URL, STREAM_ENABLED, MAX_TOKEN  # Declare globals for updates
+    session_max_token = 0
     if not API_KEY:
         console.print("[bold red]API key not found in database. Please set it with '/config api'.[/]")
         # Prompt for API key on startup if missing
@@ -145,7 +146,6 @@ def chat():
 
     # Initialize selected_model with default if available, else None (session-specific changes via /model)
     selected_model = selected_model_default  # Set from DB load, or None
-    
     client = OpenRouter(api_key=API_KEY)
     if selected_model:
         console.print("[bold blue]Welcome to oAI! Type your message, use '/help' for examples, or 'exit'/'quit'/'bye' to end.[/] \n[bold red]Active model : %s[/]" %(selected_model["name"]))
@@ -196,6 +196,26 @@ def chat():
                         console.print("[bold red]Invalid input. Enter a number.[/]")
                 continue
 
+            # Handle /maxtoken command (session limit)
+            if user_input.startswith("/maxtoken"):
+                args = user_input[10:].strip()
+                if not args:
+                    console.print(f"[bold blue]Current session max tokens: {session_max_token}[/]")
+                    continue
+                try:
+                    new_limit = int(args)
+                    if new_limit < 1:
+                        console.print("[bold red]Session token limit must be at least 1.[/]")
+                        continue
+                    if new_limit > MAX_TOKEN:
+                        console.print(f"[bold yellow]Warning: Cannot exceed stored max token limit ({MAX_TOKEN}). Capping to {MAX_TOKEN}.[/]")
+                        new_limit = MAX_TOKEN
+                    session_max_token = new_limit
+                    console.print(f"[bold green]Session max tokens set to: {session_max_token}[/]")
+                except ValueError:
+                    console.print("[bold red]Invalid token limit. Provide a positive integer.[/]")
+                continue
+
             # Handle /config command
             if user_input.startswith("/config"):
                 args = user_input[8:].strip().lower()  # Get args after "/config"
@@ -230,7 +250,30 @@ def chat():
                         console.print(f"[bold green]Streaming {'enabled' if sub_args == 'on' else 'disabled'}.[/]")
                     else:
                         console.print("[bold yellow]Usage: /config stream on|off[/]")
-                # **UPDATED:** Handle /config model for setting default only
+                # Handle /config maxtoken (global already declared at top)
+                elif args.startswith("maxtoken"):
+                    sub_args = args[9:].strip()
+                    if not sub_args:
+                        console.print(f"[bold blue]Stored max token limit: {MAX_TOKEN}[/]")
+                        continue
+                    try:
+                        new_max = int(sub_args)
+                        if new_max < 1:
+                            console.print("[bold red]Max token limit must be at least 1.[/]")
+                            continue
+                        if new_max > 1000000:  # Reasonable upper cap to prevent typos
+                            console.print("[bold yellow]Warning: Max token limit capped at 1M for safety.[/]")
+                            new_max = 1000000
+                        set_config('max_token', str(new_max))
+                        MAX_TOKEN = new_max
+                        console.print(f"[bold green]Stored max token limit updated to: {MAX_TOKEN}[/]")
+                        # Adjust session if needed
+                        if session_max_token > MAX_TOKEN:
+                            session_max_token = MAX_TOKEN
+                            console.print(f"[bold yellow]Session token limit adjusted to match: {session_max_token}[/]")
+                    except ValueError:
+                        console.print("[bold red]Invalid token limit. Provide a positive integer.[/]")
+                # Handle /config model (unchanged)
                 elif args.startswith("model"):
                     sub_args = args[6:].strip()  # After "model" (optional search term)
                     search_term = sub_args if sub_args else ""
@@ -262,7 +305,7 @@ def chat():
                         except ValueError:
                             console.print("[bold red]Invalid input. Enter a number.[/]")
                 else:
-                    # /config with no args: Display current configs.
+                    # Display all configs
                     DEFAULT_MODEL_ID = get_config('default_model')  # Load default model ID from DB
                     table = Table("Setting", "Value", show_header=True, header_style="bold magenta")
                     table.add_row("API Key", API_KEY or "[Not set]")
@@ -270,6 +313,8 @@ def chat():
                     table.add_row("Streaming", "Enabled" if STREAM_ENABLED == "on" else "Disabled")
                     table.add_row("Default Model", DEFAULT_MODEL_ID or "[Not set]")
                     table.add_row("Current Model", "[Not set]" if selected_model is None else str(selected_model["name"]))
+                    table.add_row("Max Token", str(MAX_TOKEN))
+                    table.add_row("Session Token", "[Not set]" if session_max_token == 0 else str(session_max_token))
                    
                     # Fetch and display credit info
                     credits = get_credits(API_KEY, OPENROUTER_BASE_URL)
@@ -285,6 +330,7 @@ def chat():
                     console.print(Panel(table, title="[bold green]Current Configurations[/]", title_align="left"))
                 continue
 
+
             # **Handle /credits command to display credits left**
             if user_input.lower() == "/credits":
                 credits = get_credits(API_KEY, OPENROUTER_BASE_URL)
@@ -296,8 +342,10 @@ def chat():
 
             # Handle /clear command to clear the screen
             if user_input.lower() == "/clear":
-                clear_screen()
+                os.system("clear" if os.name == "posix" else "cls")  # Cross-platform clear
                 DEFAULT_MODEL_ID = get_config('default_model')
+                token_value = session_max_token if session_max_token != 0 else " Not set"
+                console.print(f"[bold cyan]Token limits: Max= {MAX_TOKEN}, Session={token_value}[/]")
                 console.print("[bold blue]Active model[/] [bold red]%s[/]" %(DEFAULT_MODEL_ID))
                 continue
 
@@ -309,30 +357,40 @@ def chat():
                     "Select or change the current model for the session. Supports searching by name or ID (not persisted).",
                     "/model gpt\nYou: 1\n(Selects first matching model for this chat only)"
                 )
+                help_table.add_row(
+                    "/maxtoken [value]",
+                    "Set temporary session token limit (≤ stored max). View limit if no value provided.",
+                    "/maxtoken 2000\n[bold green]Session max tokens set to: 2000[/bold green]"
+                )
                 help_table.add_row(
                     "/config model [search]",
                     "Set a default model that loads on startup; displays models for selection (persisted in DB, doesn't change current session model).",
                     "/config model gpt\n(Shows GPT models; user selects to set as default)"
                 )
+                help_table.add_row(
+                    "/config maxtoken [value]",
+                    "Set stored max token limit (persisted in DB).",
+                    "/config maxtoken 50000\n[bold green]Stored max token limit updated to: 50000[/bold green]"
+                )
                 help_table.add_row(
                     "/config api",
                     "Set or update the OpenRouter API key.",
-                    "/config api\nEnter new API key: sk-...\n[bold green]API key updated and client reinitialized![/bold green]"
+                    "/config api\nEnter new API key: sk-...\n[bold green]API key updated and client reinitialized![/]"
                 )
                 help_table.add_row(
                     "/config url",
                     "Set or update the base URL for OpenRouter.",
-                    "/config url\nEnter new base URL: https://api.example.com/v1\n[bold green]Base URL updated![/bold green]"
+                    "/config url\nEnter new base URL: https://api.example.com/v1\n[bold green]Base URL updated![/]"
                 )
                 help_table.add_row(
                     "/config stream on/off",
                     "Enable or disable response streaming.",
-                    "/config stream off\n[bold green]Streaming disabled.[/bold green]"
+                    "/config stream off\n[bold green]Streaming disabled.[/]"
                 )
                 help_table.add_row(
                     "/config",
-                    "View all current configurations, including credits.",
-                    "/config\n(Displays table with credits info)"
+                    "View all current configurations, including credits and token limits.",
+                    "/config\n(Displays table with token limits & credits)"
                 )
                 help_table.add_row(
                     "/credits",
@@ -342,7 +400,7 @@ def chat():
                 help_table.add_row(
                     "/clear",
                     "Clear the terminal screen for a clean interface.",
-                    "/clear\n[bold cyan]Screen cleared. Ready for your next input![/bold cyan]"
+                    "/clear\n[bold cyan]Screen cleared. Ready for your next input![/]"
                 )
                 help_table.add_row(
                     "/help",
@@ -357,7 +415,7 @@ def chat():
                 help_table.add_row(
                     "Exiting",
                     "Quit the chat app with either 'exit', 'quit' or 'bye'",
-                    "exit\n[bold yellow]Goodbye![/bold yellow]"
+                    "exit\n[bold yellow]Goodbye![/]"
                 )
                 console.print(Panel(help_table, title="[bold cyan]oAI (Version %s) Chat Help - Command Examples[/]" %(version), title_align="center", subtitle="oAI can be found at https://iurl.no/oai", subtitle_align="center"))
                 continue
@@ -372,7 +430,7 @@ def chat():
             file_attachments = []
 
             # Regex to find @path (e.g., @/Users/user/file.jpg or @c:\folder\file.txt)
-            file_pattern = r'@([^\s]+)'  # @ followed by non spaces
+            file_pattern = r'@([^\s]+)'  # @ followed by non-spaces
             for match in re.finditer(file_pattern, user_input):
                 file_path = match.group(1)
                 expanded_path = os.path.expanduser(os.path.abspath(file_path))
@@ -416,6 +474,15 @@ def chat():
                 console.print("[bold red]Prompt cannot be empty.[/]")
                 continue
 
+            # Prepare API call params, including max_tokens if session limit is set
+            api_params = {
+                "model": selected_model["id"],
+                "messages": [{"role": "user", "content": message_content}],
+                "stream": STREAM_ENABLED == "on"
+            }
+            if session_max_token > 0:
+                api_params["max_tokens"] = session_max_token
+
             # Send to model with streaming (enable/disable based on config)
             is_streaming = STREAM_ENABLED == "on"
             if is_streaming:
@@ -424,11 +491,7 @@ def chat():
                 console.print("[bold green]Thinking...[/]", end="\r")
 
             try:
-                response = client.chat.send(
-                    model=selected_model["id"],
-                    messages=[{"role": "user", "content": message_content}],
-                    stream=is_streaming  # Enable/disable streaming
-                )
+                response = client.chat.send(**api_params)
             except Exception as e:
                 console.print(f"[bold red]Error sending request: {e}[/]")
                 continue